Mariena Quintanilla: Customer Advocacy for Data Products, and AI in Customer Operations

April 1, 2024

Summary

Hello everyone. My name is Chun Jiang, CEO and co-founder of Monterey AI. Monterey AI helps companies make it super easy to aggregate triaged and analyze user feedback from sources like support tickets, sales, transcript, and so many more. So the team can always keep a pulse on customers' voices. Make users happy and the unlock, the next growth opportunity. Today, I am super excited to welcome you to our podcast. Making waves. Where we feature leaders in a voice of a customer space to share their stories, practice, and insights.

Today, our guest is Mariena Quintanilla. Formerly a Foursquare and Factual. And the leader in client and sales solutions engineering. Who immerses herself in customer operations daily. And the call herself a data nerd. With her engineering background, we dive into you the fascinating aspects of customer advocacy in the data product world. How do you navigate organizational systems, how to manage customer expectations and AI and accessibility.

If you like our podcast and the content, don't forget to subscribe and follow us on monterey.ai.

Without further ado. That's welcome Mariena.

If you enjoy this podcast, don’t forget to subscribe on YouTube, of follow us on Monterey AI.


Speakers

Where to find Mariena Quintanilla:

• LinkedIn: https://www.linkedin.com/in/mariena/

Where to find Chun Jiang:

• LinkedIn: https://www.linkedin.com/in/chunonline

• Website: https://www.monterey.ai

Transcripts

Chun:

Hello everyone. My name is Chun Jiang, CEO and co-founder of Monterey AI. Monterey AI helps companies make it super easy to aggregate triaged and analyze user feedback from sources like support tickets, sales, transcript, and so many more. So the team can always keep a pulse on customers' voices. Make users happy and the unlock, the next growth opportunity. Today, I am super excited to welcome you to our podcast. Making waves. Where we feature leaders in a voice of a customer space to share their stories, practice, and insights.

Today, our guest is Mariena Quintanilla. Formerly a Foursquare and Factual. And the leader in client and sales solutions engineering. Who immerses herself in customer operations daily. And the call herself a data nerd. With her engineering background, we dive into you the fascinating aspects of customer advocacy in the data product world. How do you navigate organizational systems, how to manage customer expectations and AI and accessibility.

If you like our podcast and the content, don't forget to subscribe and follow us on monterey.ai.

Without further ado. That's welcome Mariena.

Mariena:

Sure, thanks for having me trying. I'm super excited that you guys have started this podcast and I'm excited to talk today. So I, you know, I've been working in tech for about 17 years and have been a little bit all over the place. Bit of a zigzaggy career. I've just been always someone who has.

Really broad interests and likes learning about everything. And I, I knew from early days that I didn't want to just be in software engineering. I wanted to really understand businesses as a whole. And so, you know, I got into product, I got into you know, the customer and like go to market side, building customer facing engineering teams.

Specifically within ad tech and location and pretty early on in my career, I started working with data. And that's really where I, I fell in love and was like, I know I want to work with data and machine learning and it's been so much fun. And I've just really enjoyed helping customers, especially get to use data products and get the value from machine learning and from a that that's what really excites me.

One of the CEOs Gil Alba is actually a four square used to talk about it as democratizing tech. And that really resonated with me. So that's a little bit about me what I care about. I'm here in Los Angeles, originally from outside of Seattle, Pacific Northwest. So always up for hiking or just hanging out with trees and anything really green.

 Chun:

I love it. I love it. Your career trajectory or like experience is super, super interesting. I mean, I'm on kind of the same boat of every time, like, I'm so interested in all the specs of like, business, like, from product design, and then product, and then now customer operations and data analytics in general I would love to hear a little bit more about the original story, like, Walk us through how you find, like, fell in love with a different spaces or different, like, very different roles, right?

In the tech industry and how you find that kind of sweet spot, like, do everything.

 Mariena:

And I mean, and it wasn't, you know, I didn't necessarily get it right away. I think it took trying some things to figure out sort of what I was looking for. So I'll start with sort of the functional side. So I was in software engineering doing application development and after six years of that, you know, wanted to try something else.

And my manager at the time really believed in me and she gave me a shot. She took a risk on me and moved me. From an engineering manager role. I went back into an IC role as a data quality engineer. And that, that was really fun to try something completely different. It was a really hard role because it was a little bit program management trying to influence VPs who are architects, who are product managers.

And I was maybe, I don't know, my early twenties at the time. And so I was surrounded by people with 10 plus years more experience with me, which was such a cool opportunity to learn and be exposed to the way they were thinking and making decisions. And I was mentored essentially by a team from Motorola.

That had come out of you see Chicago I think Urbana Champaign, and they were just true experts on data quality. And they mentored me and sort of trained me on how to think about data quality data quality, best practices and principles. And I became the subject matter expert for Advertising systems at Yahoo, and because I already had 6 years of experience in that business unit, I really understood our advertising systems pretty well.

And then they gave me that sort of functional context of how to now approach it from a data quality lens. And it was. Really exciting to be able to follow the data from ad serving all the way through it getting sort of logged and ingested and processed by data systems and then through billing and reporting.

And I think part of what really made it interesting to me was that I grew up with a family of storytellers, and I started to see that within data, there were just stories all around us, and it was really cool to try to find the stories and the data, and really create that narrative and that, that just, I don't know, it still gets me excited today, you can, if you're on video, anyone watching on video, you can see like, I'm just, I'm excited probably hear it in my voice.

And from the data quality lens, the sort of ability of then good or bad data to really change that narrative was something I saw really early on and I kind of likened it to there could be potholes throughout this sort of road, right? That the data, the journey of the data was taking or even bigger than potholes.

You know, all kinds of road issues that might close down even a road. So I think that helped me see the importance of data quality and the need to really look at it early on, but really throughout that journey. And that's kind of where I fell in love with data.

The second half of your question in terms of functions I wanted to get closer to the customer to be honest.

And so from engineering, I moved into product management and I, I had known and worked with a lot of product managers and I thought that was going to be the Holy grail for me. And four years into product management, I realized I missed being closer to the technology. I wasn't as close to the customer as I thought I would be.

You're really being pulled in so many directions and products and it was incredibly hard role. I really commend anyone in response. So much respect for product managers because it's, it's so hard. You're pulled in so many different directions. And so from there I looked for a role that would be more customer facing and more technical.

I didn't know what that would be. I just started talking to people in my network and kind of talk. Here's what I'm good at. What do you think I should do? What do you know of? And I landed in solutions architecture and this space that is really the intersection of product, of technology, of customers has really been something that I fell in love with.

And so I've spent the last seven years there, both in pre and post sales role, sales engineering support engineering. And it's, it's such a fun time because you really get to see how customers you get to see how customers, the products impact them and you directly solve their problems, which is really exciting.

 Chun:

Love it. I love it. It's still a lot of, lot of. Responsible responsibility to untangle there. It's funny when I was thinking about hiring one sales engineer this morning, I'm like, well, this sales engineer also need to do this kind of customer engineer. This need to do a lot of growth engineering. You need to do a product engineering. And I was like trying to label you to see, okay, what does Marina know? How do we find someone who's like her? And the, one of the biggest challenge, especially like in the field of like AI or this days is everything is getting like even more complicated from my data perspective.  

And what I find, at least like when we talk with customers or onboard customers to Monterey AI We found that most of the time we are not only teaching customers how to use the tool, but also to help them understand, like, what does like data look like in their organization? Where do they see the most available data? Is it in like Salesforce? Is it in like send desk? What are, what is the cadence they're looking at? What are the primary metrics that they're trying to track there? Okay. A lot of this are like data educations  to, to some sense, and that AI education and then tool education,

​ I would love to just learn more about, like, the onboarding secret you have, like, what is the best way to think from, like, customer perspective, but also really smoothly. Like onboarding into this, like, massive of data knowledge and a knowledge to some sense.

 Mariena:

I think you know, that's a, it's a hard question to answer and it's definitely not one size fits all, but I think one of the things I learned when I was in solutions architecture was I think really that value of the discovery phase. And, you know, the, the company I was working for was a marketing analytics platform that was, you know, Taking data in from customers, all the advertising data and then building a machine learning models on top of that to help them understand how their media was performing.

And basically, you know, whether they should buy on Facebook or spend more on TV, things like that. But what what I learned through that experience was really the importance of that discovery phase ideally before you even. Pitch a customer on exactly what you're selling to them, depending on, I guess, the complexity of the offering.

But it was really needing to actually understand. And dive deep into their data. And so that's what I would do is I would leave these discovery calls and kind of came up with a template that would really help me guide those conversations to understand. What do they know? And what do they know? Not know about their own data, right?

Do they have blind spots? Because you're really trying to assess the maturity of the organization because, you know, If it's a immature organization, then that's a really strong signal that they don't have good data hygiene and you're going to run into all kinds of issues. And so, you know, pretty early on that, well, then that just means higher costs, right?

More time, more money. And so you, you kind of need to either bake that in or you need to make the, Decisions in your own organization about are those the customers we want to work with, especially if you're a small startup, right? And you might need to focus on the customers where your product is going to be successful.

So I feel like there, there's definitely one angle there about you know, choosing which customers you're going to work with that have the height for your product until you start to develop maybe more mature. ingestion processes, right? That can themselves maybe do some of the cleaning and deal with the really common issues because no one has perfect data.

So you also still then need to develop different either tooling or start to understand the types of patterns, you know, that your product sort of requires to be able to really find and get rid of that bad data that maybe all customers have certain types of data issues. So that takes time though.

Chun:

It definitely  takes time. Voice of a customers, this kind of concept, like. In other organizations or in non data or analytics products is already very complicated. It takes a lot of like, you know, stakeholders to move the needle there. But I would be curious, like say mission about like after onboarding, right after the customers are like using the platform after they start seeing the data a lot of like voice of customer, like feedback not only really related to the tool they're using or would provide. But mostly, like, this year, this kind of ongoing discovery, ongoing, like, education about their pipe data pipeline, their data hygiene, ​ I would love to hear a little bit more from, like, your experience about that perspective.

 Mariena:

Yeah, I think you know, and the products I've worked on have been kind of a mix of data coming in that isn't from customers that then is being used to create the data products sold to the customers. And in that case, the issues are a bit different from when you're ingesting, say, in that marketing use case, ingesting data, and then giving the customers a product on top of their 1st party data.

Things like that, but I, I think in terms of ongoing education for, for them to clean the data, one of the things that I think has been helpful is really how do you expose some of the metrics to them that can help them be aware of, things that are going to influence their, their ability to really give value from the product, right?

And so it's almost like you know, within engineering, we have, we talk about like health checks and we have different, you know, monitoring and dashboards that we use as engineers and with product managers to sort of assess the health of the product, but different from maybe the metrics you're sharing with executives and senior leaders for, you know, health.

The, the business health, right? But a lot of those are sort of leading metrics or sort of secondary metrics that I think can, can help detect issues before they come up and they better help us understand what's driving maybe those other metrics. Right. And so I, sometimes it's helpful to even have pages within your products that help customers see some of those things.

And, you know, maybe your front page, it can redirect and say, Hey, some of these things we're seeing that, you know, they're going to. It's like the warning warning signs, right? That, you know, if you fix this, you know, you're going to be able to get more value out of, you know, maybe this feature or that feature, things like that.

 Chun:

so it's a lot of, like, if then kind of scenario play there for the customers.

 Mariena:

I mean, potentially, I think it can also be depending on the maybe machine learning base. I don't know if it's heuristics or machine learning, but the types of, you know, you, you tend to know some of the different types of data issues that you're fixing during onboarding. So sometimes it's the setting up checks for those types of things and integrating the same type of data QA that you might do during onboarding.

Into like your ongoing processes and it's going to be phased, right? You might do it one time during onboarding as sort of your first bar and then start to work up to how do we monitor for some of these things so we can send the customer an email or have the account manager reach out when these things go off.

And then eventually longer term, get more towards that self serve Nirvana state of proactive, right? I think they call it like proactive customer support, right? Where you can just reach out right away before. Who affects them?

Chun: 

I love it. What has been, what was the most like challenging? I would love to hear like one story or just like one example about a customer came to you with a problem or maybe with a solution suggestion that is like totally not actually the root cause which happens a lot. I think like in customer's life. Operation Door Success is like, okay, customers, they are super loved, they're friendly, they wanna tell you like how to solve the problem, but sometimes, especially in like highly complicated platform, that is not actually the root cause. I would love to hear a bit more from interacting with like customers. Like how do you get to the next step to find the root causes? For data products.

 Mariena:

Yeah. Yeah. I think for data products, sometimes you really do have to go sort of multiple levels and understand what's the source of that data issue. When we were, you know, originally I had led an effort at Foursquare to classify all of our issues by root cause and the sort of more traditional, I would say software engineering root causes you might think about, was this related to the ability of the tech to, you know, scale.

Was it a bug in the software, you know, things like that, or some infrastructure issue or a user issue, right? Those didn't necessarily translate directly often to when we were trying to classify the root cause for our data, the data licensing product. Instead, we had to almost look at, okay, where did this data issue come from?

And was it a problem of not having enough source data?

 Chun:

yeah.

 Mariena:

Was it a problem of conflicting data? Things like that. And so we almost had to come up with a whole different taxonomy when we were looking at that root cause. And for sure the customers, like, I mean, even the software ones, they're not going to understand or talk to the root cause of those types of issues, but for data even more so.

And so we, we really just had to come up with a whole new taxonomy. And to do that, we had to go sort of one by one through, you know, a sample. Of of tickets and really work very cross functionally with product and with the engineering team and think about not just I would say the root cause, but what is it that will help us take action?

So, for instance, by tracking and understanding if the issue was the quality of the data we were ingesting versus not having enough data. That could inform the business development team for how to approach our partnerships, right? Cause in this case, we were ingesting third party data. So I think you also have to look at it from that lens of what's going to help you take action and not just sort of classifying things in the way that maybe feels like this is sort of, I don't know, the most accurate.

 Chun:

Yeah. I like it. I like it. The triage system, like on our platform, it came from a design partner, it's from very early days. So then that originally from one of the problem, we find that like, okay, the taxonomy that customers have in their mind are very different than the taxonomy, like organizations have in their mind. And then first time for most of company, like when we were presenting this triage, you're like, oh. Is this kind of like just a labeling system for like customer issues? I'm like, no, like, think of this as more like a routing system internally. You can use this like triage system to like, triage, triage to different product errors or triage to like different root causes to triage to different, like, initiatives. It totally depends on like, what kind of a taxonomy is the best one to, for you to take action.

 Mariena:

Yeah.

 Chun:

I definitely love that. As we both know, it's probably everyone know like for it takes time and takes like resources from engineering and sales and product to really and partnership to really like improve problems like data quality or scale of data.

This whole process, like, how do you, how do you keep the relationship with that customers? How do you make them like, understand this kind of resources constraint there, but at the same time, like, not discourage them to give you more feedback.

 Mariena:

yeah, I think managing expectations is really important in that as well as just keeping an open line of communication. I think with data products, especially and licensing data, I think we've all think about it as partnerships to like, we don't want to just be in a vendor relationship where it's, you know, You know, sort of one sided.

We're giving you this thing, but rather we're helping each other improve. And so we're helping you build a product with this data and you're giving us feedback so that now we can have this cycle of improvement of the data itself. Talking about managing expectations, I think a piece of it comes in where when you're getting that feedback, it's it's understanding the context behind it.

Right? Sometimes you're getting feedback. That is very sort of just backhanded. Like, we see this. We see that. And they're they're just giving it to you because they see it. They don't necessarily even have an urgency or a strong need to see it resolved. And you have to really distinguish that kind of feedback from feedback that is, hey, this is blocking us, you know, really getting value from that data, right?

And having. Building a valuable product. And so that you can actually prioritize that. That's one piece of it. And then once you even know that, though, you have to internally figure out how do you deal with that feedback and what kind of, you know, effort is going to go into You know, doing something with that feedback.

If your organization isn't really set up to react quickly, then customers need to know that if it's going to be, you know, we'll come back in a quarter or we review these things quarterly, whatever the organization is set up to do, and I wouldn't say there's one right answer, but the one thing you do have to do is you have to talk to your customers, manage those expectations so they know either when you're going to review the feedback.

As well as once you've reviewed it, what are you doing with that feedback? Right.

 Chun:

yeah, yeah. That reminds me of the LinkedIn post I posted about last week with all the AI customer support bot you will see this, like, really, really long, just, like, totally useless, like, conversation that are trying to make it a little more personalized, but trying to scale their like customer communication, but from customer side, it's like, it's not telling me anything about like timeline is not giving me any, like, context. So I would say that's probably 1 opportunity, like, for AI to look into it or improve.

 Mariena:

Yeah, yeah. And I think that's been the hard part is doing that. Well, I think you're, you're only really getting that level of relationship management with an enterprise customer success. You're definitely not going to get it from a customer support agent because of the volume of tickets with, and they also maybe don't have, they're not as close.

To the process to get that information, maybe from product and engineering, for instance, so definitely interesting opportunities for AI to make that whole process more efficient because even on the enterprise side, when you have the people doing that, as I discussed, it's a pretty people heavy process and it can be pretty disruptive to the organization.

 Chun:

Yeah, yeah, yeah. I love that. I love that. It's a question. Another interesting question from my own experience, like, onboarding, like enterprise customers. We mentioned about, like, the onboarding effort of like data products and a lot of time is about, like. Data literacy education, and it's really hard to, like, scale that during, like, onboarding, because you're talking with, like, people probably, like, one by one sometimes, right? What are some, like, ideas or practice that you've done before to help, like, customers understand architecture of the data? Or how to use like data for storytelling in a short time or how to read the data to

 Mariena:

well, I mean, in the analytics product that we had, I think the key was to start for at least productizing it from a SAS perspective, start to identify the different types of stories that come up in the data and almost look for the stories for them through templates. And then surface those, right.

And ideally you're moving away from kind of just modules like reports and you are able to, I think, create the stories. But I think there's a lot of opportunities for generative AI to, you know, help with, help with that as well. But I think if kind of extracting some of the different types of things that come up and then sort of looking for, for this customer, is this one applicable or not

And  working with AI to really. You know, showcase that to them. I think, I think it's going to lower the bar, right. You know, in the past sort of larger organizations have been able to do this because they've had the resources to have, you know, the sort of data analysis, data analysts, and maybe business analysts or different combinations of roles that could put that story together and support like a senior leader, for instance, or

Chun:

Yeah.

Mariena:

You know, that product manager, and now it's hopefully going to make a lot of this more accessible.

 Chun:

Yeah. Yeah. Yeah. One of the my favorite features of our chatbot is like, we don't give you a blank bot, like blank paper to start with that. We give you kind of suggested questions based on the data we see in your workspace. And I'm going to say, like, I think most of folks would like, start with, like, suggest other questions first, and then kind of going through this process.

Okay, maybe I should ask, like, a gentleman areas questions there. Awesome. Awesome. Let's see. We talk a lot about the customer onboarding. We talk a lot about the you know, manage expectations. I want to dive a little bit deeper into the process. Okay. That you have helped like organization come up for better kind of like a customer operations there. I know. Well, I would just like you to tell the story. Tell me a little bit more about, like, you know, how you help organizations understand the importance of, like, setting up this, like, really good, like, triage, like a routing of customer operations and how you, like, Yeah. Get like a product and the engineering and the sales team on the same page about, hey, this is what we need to do on like customer management and the customer communication site.

Mariena:

Yeah, I think in terms of support, especially some of the organizations that I've been in have not, when I've kind of entered, they haven't had strong processes, if any, that are well defined for how we deal with issues. And so, you know, these have been more like enterprise SAS organizations and, you know, you might have a customer success manager engaging with.

Someone on a team like I've had, who's sort of their, their technical person to help figure out like what's actually happening and route it and, you know, figure out, okay, no, it needs to go to this engineering team, things like that. But those don't, it doesn't scale at all and it's not measurable. And at the end of the day, we also often haven't actually had even that formalized like support engineering role, for instance, or you're directly going to engineering teams, it gets very chaotic.

And so. Some of the things that I've introduced, you know, some of them are just best practices like having a central ticketing system for internal issues to get raised. And then from there, that's where you can then start to actually really leverage continuous improvement, right? To just. Start to knock things out sort of 1 by 1 really based on whatever is going to be most impactful in your organization.

And I think for me, I'm I'm talking at a minimum at a quarterly level to various stakeholders to both, just get a sense of what is the most impactful. Is it focusing on how we triage? Is it focusing on how we move issues? We can't resolve into the engineering teams. Is it transparency and reporting? So I think that conversation it's good to always.

Keep your nose on that so that you can be adjusting, you know, just like product managers are adjusting roadmaps on the customer support side to like using those same principles of like roadmap planning and sprints, I think has been helpful to create like that culture of continuous improvement.

And instead of coding, sometimes we're working more on process related stuff or. Or maybe we're building small tools. And then some of the, the other things that have been impactful has been like actually creating like routing systems you were talking about. Your, your guys's system has this sort of concept of, of routing through different labels, but you know, we've done that as well.

Introducing SLA is a lot of these organizations didn't have SLAs for support service level agreements, or I guess objectives maybe if they're not. No contractual, I think that goes a long way to getting alignment between the technical product and customer facing teams around when, like, what levels of responsiveness are needed for what levels of issues so that everything doesn't feel like a fire and you're not, you know, sort of.

Really just draining everyone and burning everyone out. So that's me is always the lens. I'm going through with a lot of these process improvements. It's not process for processes sake, but it's really about what processes will help different teams in the organization work better together. So there's less friction.

 Chun:

Yeah.

 Mariena:

We can manage expectations better. And so that the, the people working on these types of issues are actually enjoying their job. They're having fun. They're learning, you know, we're, we're not just kind of burning people out because that can easily happen. And some of these sort of technical support solution roles as well You know, the data, the data rolls.

 Chun:

Yeah. Everyone love like serving customers. I mean, it's also another interesting case to see, like, what is the best way to like phrase customer feedback to motivate the team to work on? We have this like, interesting observation that, our categorization of appreciation. So we can like auto detect, like, what are the things that people are saying to love about your product? And then if we  connect that and send that to like Slack channel engineers, I feel super motivated that when I see the data, I come in.

 Mariena:

It's interesting you bring that up because one of the things that Foursquare did when I was there that I really liked is they would do regular, like pulse surveys so that as people managers, we could understand you know, and. Corrective need be where things weren't working well and what my team always scored the highest sometimes 100%.

I had multiple I think months and quarters where it was 100 percent on the question. Do you understand the value of your work and how it relates to customers or something to that effect? And I always thought it was really cool. Cool. And it, to me, it was also very intuitive, right? We are the ones that were the glue between the product and the customers.

We were the ones really making sure any last mile solutions, you know, got delivered or that they understood things that might keep them from getting value in terms of like that technical education. So I think anything that you can do to help other people who aren't as close to the customer and get that kind of feedback,

 ​It helps them understand how their work is helping people.

I mean, that's why I got into these roles because that's what I wanted. And I wanted to make sure I was working on products that were actually solving problems for customers and get to see that firsthand. And it's a, it's a really fun place to work, but I love that you're finding ways to help other people get that joy as well. So they can also experience

Chun:

I love  that. I love that. I love that. I don't think it's many like organization that are realizing this because we spend so much time like organization spend so much time and money trying to boost the morale there. But by the end of day, like, the magic and power happens. at like the point of we can connect customers with builders and that's it.

 Mariena:

Yeah. I think there's a lot of software engineers, especially that get into that. Career because they enjoy solving problems inherent, right? And so when you can see that something you've done solved a problem for someone that's really satisfying, I mean, obviously we all get motivated by different things, but I think for a lot of people that in of itself, it makes you feel like what you worked on matters.

 Chun:

I love it. I love it. But that half hour went really fast. All right. I don't want to take too much of a time, but I would love to kind of like end on a question that asks everyone what has been one or two like books that you read, read like recently that has been like really impactful and you. Love more people to read it.

 Mariena:

Yeah, well, it's, you know, it's not necessarily technical or related exactly to what we're talking about, but the book I read in the fall that was really impactful for me was Braiding Sweetgrass, and it's written by an indigenous woman who's also a A researcher, she's an ecologist and teaches ecology university level and she, the whole book is about that intersection of science and her indigenous culture and how they approach the relationship with the earth and.

then the scientific view of the earth. And so she talks about data. She talks about the scientific method and experimentation and, you know, how we approach those things with that scientific mind and the sort of other side of that. And it's, it's really cool to see that intersection and open my eyes. So I highly recommend that book to anyone.

Breeding, Breeding Sweetgrass.

 Chun:

Awesome. Awesome. It's also very fitting to the spring theme.

 Mariena:

Yes. And I mean, it does have, you know some context in the world of LLMs, right? We're, we're using a lot more energy as we run large language models. And so a lot of the things she's talking about with our relationship to the Earth and things like that, right? There's, there's a lot there.

Definitely a through line there. So highly recommend it.

Chun: 

I love it. I love it. Well, thank you so much. Is there anything else I haven't asked you and you want to add?

Mariena:

 Yeah, I mean, so my actually most recent venture, I, I switched gears a little bit. So I talked in the beginning about democratizing tech, and that was something that for me has been so important that I actually, on my side, I've started my own business now, Melonhead, and really want to focus on helping individuals as well as different organizations really, you Get the value of all these different innovations in tech that are happening.

So generative AI is what I'm focusing on right now, doing both community focused workshops for individuals to help them really feel more empowered and less from a place of fear with generative AI, and then for organizations as well, helping them actually deploy. Generative AI solutions in their organization.

So hopefully between both of those, you know, we'll see the really impact of generative AI and its potential really trickle down to, you know, all, all kinds of people in society is sort of my goal.

 Chun:

Awesome. Well, again, thank you so much. We, I really enjoy it. I got conversation. I already like learned a lot again from you. And yeah, and thank you everyone for time and we will see you next time.

Summary

Hello everyone. My name is Chun Jiang, CEO and co-founder of Monterey AI. Monterey AI helps companies make it super easy to aggregate triaged and analyze user feedback from sources like support tickets, sales, transcript, and so many more. So the team can always keep a pulse on customers' voices. Make users happy and the unlock, the next growth opportunity. Today, I am super excited to welcome you to our podcast. Making waves. Where we feature leaders in a voice of a customer space to share their stories, practice, and insights.

Today, our guest is Mariena Quintanilla. Formerly a Foursquare and Factual. And the leader in client and sales solutions engineering. Who immerses herself in customer operations daily. And the call herself a data nerd. With her engineering background, we dive into you the fascinating aspects of customer advocacy in the data product world. How do you navigate organizational systems, how to manage customer expectations and AI and accessibility.

If you like our podcast and the content, don't forget to subscribe and follow us on monterey.ai.

Without further ado. That's welcome Mariena.

If you enjoy this podcast, don’t forget to subscribe on YouTube, of follow us on Monterey AI.


Speakers

Where to find Mariena Quintanilla:

• LinkedIn: https://www.linkedin.com/in/mariena/

Where to find Chun Jiang:

• LinkedIn: https://www.linkedin.com/in/chunonline

• Website: https://www.monterey.ai

Transcripts

Chun:

Hello everyone. My name is Chun Jiang, CEO and co-founder of Monterey AI. Monterey AI helps companies make it super easy to aggregate triaged and analyze user feedback from sources like support tickets, sales, transcript, and so many more. So the team can always keep a pulse on customers' voices. Make users happy and the unlock, the next growth opportunity. Today, I am super excited to welcome you to our podcast. Making waves. Where we feature leaders in a voice of a customer space to share their stories, practice, and insights.

Today, our guest is Mariena Quintanilla. Formerly a Foursquare and Factual. And the leader in client and sales solutions engineering. Who immerses herself in customer operations daily. And the call herself a data nerd. With her engineering background, we dive into you the fascinating aspects of customer advocacy in the data product world. How do you navigate organizational systems, how to manage customer expectations and AI and accessibility.

If you like our podcast and the content, don't forget to subscribe and follow us on monterey.ai.

Without further ado. That's welcome Mariena.

Mariena:

Sure, thanks for having me trying. I'm super excited that you guys have started this podcast and I'm excited to talk today. So I, you know, I've been working in tech for about 17 years and have been a little bit all over the place. Bit of a zigzaggy career. I've just been always someone who has.

Really broad interests and likes learning about everything. And I, I knew from early days that I didn't want to just be in software engineering. I wanted to really understand businesses as a whole. And so, you know, I got into product, I got into you know, the customer and like go to market side, building customer facing engineering teams.

Specifically within ad tech and location and pretty early on in my career, I started working with data. And that's really where I, I fell in love and was like, I know I want to work with data and machine learning and it's been so much fun. And I've just really enjoyed helping customers, especially get to use data products and get the value from machine learning and from a that that's what really excites me.

One of the CEOs Gil Alba is actually a four square used to talk about it as democratizing tech. And that really resonated with me. So that's a little bit about me what I care about. I'm here in Los Angeles, originally from outside of Seattle, Pacific Northwest. So always up for hiking or just hanging out with trees and anything really green.

 Chun:

I love it. I love it. Your career trajectory or like experience is super, super interesting. I mean, I'm on kind of the same boat of every time, like, I'm so interested in all the specs of like, business, like, from product design, and then product, and then now customer operations and data analytics in general I would love to hear a little bit more about the original story, like, Walk us through how you find, like, fell in love with a different spaces or different, like, very different roles, right?

In the tech industry and how you find that kind of sweet spot, like, do everything.

 Mariena:

And I mean, and it wasn't, you know, I didn't necessarily get it right away. I think it took trying some things to figure out sort of what I was looking for. So I'll start with sort of the functional side. So I was in software engineering doing application development and after six years of that, you know, wanted to try something else.

And my manager at the time really believed in me and she gave me a shot. She took a risk on me and moved me. From an engineering manager role. I went back into an IC role as a data quality engineer. And that, that was really fun to try something completely different. It was a really hard role because it was a little bit program management trying to influence VPs who are architects, who are product managers.

And I was maybe, I don't know, my early twenties at the time. And so I was surrounded by people with 10 plus years more experience with me, which was such a cool opportunity to learn and be exposed to the way they were thinking and making decisions. And I was mentored essentially by a team from Motorola.

That had come out of you see Chicago I think Urbana Champaign, and they were just true experts on data quality. And they mentored me and sort of trained me on how to think about data quality data quality, best practices and principles. And I became the subject matter expert for Advertising systems at Yahoo, and because I already had 6 years of experience in that business unit, I really understood our advertising systems pretty well.

And then they gave me that sort of functional context of how to now approach it from a data quality lens. And it was. Really exciting to be able to follow the data from ad serving all the way through it getting sort of logged and ingested and processed by data systems and then through billing and reporting.

And I think part of what really made it interesting to me was that I grew up with a family of storytellers, and I started to see that within data, there were just stories all around us, and it was really cool to try to find the stories and the data, and really create that narrative and that, that just, I don't know, it still gets me excited today, you can, if you're on video, anyone watching on video, you can see like, I'm just, I'm excited probably hear it in my voice.

And from the data quality lens, the sort of ability of then good or bad data to really change that narrative was something I saw really early on and I kind of likened it to there could be potholes throughout this sort of road, right? That the data, the journey of the data was taking or even bigger than potholes.

You know, all kinds of road issues that might close down even a road. So I think that helped me see the importance of data quality and the need to really look at it early on, but really throughout that journey. And that's kind of where I fell in love with data.

The second half of your question in terms of functions I wanted to get closer to the customer to be honest.

And so from engineering, I moved into product management and I, I had known and worked with a lot of product managers and I thought that was going to be the Holy grail for me. And four years into product management, I realized I missed being closer to the technology. I wasn't as close to the customer as I thought I would be.

You're really being pulled in so many directions and products and it was incredibly hard role. I really commend anyone in response. So much respect for product managers because it's, it's so hard. You're pulled in so many different directions. And so from there I looked for a role that would be more customer facing and more technical.

I didn't know what that would be. I just started talking to people in my network and kind of talk. Here's what I'm good at. What do you think I should do? What do you know of? And I landed in solutions architecture and this space that is really the intersection of product, of technology, of customers has really been something that I fell in love with.

And so I've spent the last seven years there, both in pre and post sales role, sales engineering support engineering. And it's, it's such a fun time because you really get to see how customers you get to see how customers, the products impact them and you directly solve their problems, which is really exciting.

 Chun:

Love it. I love it. It's still a lot of, lot of. Responsible responsibility to untangle there. It's funny when I was thinking about hiring one sales engineer this morning, I'm like, well, this sales engineer also need to do this kind of customer engineer. This need to do a lot of growth engineering. You need to do a product engineering. And I was like trying to label you to see, okay, what does Marina know? How do we find someone who's like her? And the, one of the biggest challenge, especially like in the field of like AI or this days is everything is getting like even more complicated from my data perspective.  

And what I find, at least like when we talk with customers or onboard customers to Monterey AI We found that most of the time we are not only teaching customers how to use the tool, but also to help them understand, like, what does like data look like in their organization? Where do they see the most available data? Is it in like Salesforce? Is it in like send desk? What are, what is the cadence they're looking at? What are the primary metrics that they're trying to track there? Okay. A lot of this are like data educations  to, to some sense, and that AI education and then tool education,

​ I would love to just learn more about, like, the onboarding secret you have, like, what is the best way to think from, like, customer perspective, but also really smoothly. Like onboarding into this, like, massive of data knowledge and a knowledge to some sense.

 Mariena:

I think you know, that's a, it's a hard question to answer and it's definitely not one size fits all, but I think one of the things I learned when I was in solutions architecture was I think really that value of the discovery phase. And, you know, the, the company I was working for was a marketing analytics platform that was, you know, Taking data in from customers, all the advertising data and then building a machine learning models on top of that to help them understand how their media was performing.

And basically, you know, whether they should buy on Facebook or spend more on TV, things like that. But what what I learned through that experience was really the importance of that discovery phase ideally before you even. Pitch a customer on exactly what you're selling to them, depending on, I guess, the complexity of the offering.

But it was really needing to actually understand. And dive deep into their data. And so that's what I would do is I would leave these discovery calls and kind of came up with a template that would really help me guide those conversations to understand. What do they know? And what do they know? Not know about their own data, right?

Do they have blind spots? Because you're really trying to assess the maturity of the organization because, you know, If it's a immature organization, then that's a really strong signal that they don't have good data hygiene and you're going to run into all kinds of issues. And so, you know, pretty early on that, well, then that just means higher costs, right?

More time, more money. And so you, you kind of need to either bake that in or you need to make the, Decisions in your own organization about are those the customers we want to work with, especially if you're a small startup, right? And you might need to focus on the customers where your product is going to be successful.

So I feel like there, there's definitely one angle there about you know, choosing which customers you're going to work with that have the height for your product until you start to develop maybe more mature. ingestion processes, right? That can themselves maybe do some of the cleaning and deal with the really common issues because no one has perfect data.

So you also still then need to develop different either tooling or start to understand the types of patterns, you know, that your product sort of requires to be able to really find and get rid of that bad data that maybe all customers have certain types of data issues. So that takes time though.

Chun:

It definitely  takes time. Voice of a customers, this kind of concept, like. In other organizations or in non data or analytics products is already very complicated. It takes a lot of like, you know, stakeholders to move the needle there. But I would be curious, like say mission about like after onboarding, right after the customers are like using the platform after they start seeing the data a lot of like voice of customer, like feedback not only really related to the tool they're using or would provide. But mostly, like, this year, this kind of ongoing discovery, ongoing, like, education about their pipe data pipeline, their data hygiene, ​ I would love to hear a little bit more from, like, your experience about that perspective.

 Mariena:

Yeah, I think you know, and the products I've worked on have been kind of a mix of data coming in that isn't from customers that then is being used to create the data products sold to the customers. And in that case, the issues are a bit different from when you're ingesting, say, in that marketing use case, ingesting data, and then giving the customers a product on top of their 1st party data.

Things like that, but I, I think in terms of ongoing education for, for them to clean the data, one of the things that I think has been helpful is really how do you expose some of the metrics to them that can help them be aware of, things that are going to influence their, their ability to really give value from the product, right?

And so it's almost like you know, within engineering, we have, we talk about like health checks and we have different, you know, monitoring and dashboards that we use as engineers and with product managers to sort of assess the health of the product, but different from maybe the metrics you're sharing with executives and senior leaders for, you know, health.

The, the business health, right? But a lot of those are sort of leading metrics or sort of secondary metrics that I think can, can help detect issues before they come up and they better help us understand what's driving maybe those other metrics. Right. And so I, sometimes it's helpful to even have pages within your products that help customers see some of those things.

And, you know, maybe your front page, it can redirect and say, Hey, some of these things we're seeing that, you know, they're going to. It's like the warning warning signs, right? That, you know, if you fix this, you know, you're going to be able to get more value out of, you know, maybe this feature or that feature, things like that.

 Chun:

so it's a lot of, like, if then kind of scenario play there for the customers.

 Mariena:

I mean, potentially, I think it can also be depending on the maybe machine learning base. I don't know if it's heuristics or machine learning, but the types of, you know, you, you tend to know some of the different types of data issues that you're fixing during onboarding. So sometimes it's the setting up checks for those types of things and integrating the same type of data QA that you might do during onboarding.

Into like your ongoing processes and it's going to be phased, right? You might do it one time during onboarding as sort of your first bar and then start to work up to how do we monitor for some of these things so we can send the customer an email or have the account manager reach out when these things go off.

And then eventually longer term, get more towards that self serve Nirvana state of proactive, right? I think they call it like proactive customer support, right? Where you can just reach out right away before. Who affects them?

Chun: 

I love it. What has been, what was the most like challenging? I would love to hear like one story or just like one example about a customer came to you with a problem or maybe with a solution suggestion that is like totally not actually the root cause which happens a lot. I think like in customer's life. Operation Door Success is like, okay, customers, they are super loved, they're friendly, they wanna tell you like how to solve the problem, but sometimes, especially in like highly complicated platform, that is not actually the root cause. I would love to hear a bit more from interacting with like customers. Like how do you get to the next step to find the root causes? For data products.

 Mariena:

Yeah. Yeah. I think for data products, sometimes you really do have to go sort of multiple levels and understand what's the source of that data issue. When we were, you know, originally I had led an effort at Foursquare to classify all of our issues by root cause and the sort of more traditional, I would say software engineering root causes you might think about, was this related to the ability of the tech to, you know, scale.

Was it a bug in the software, you know, things like that, or some infrastructure issue or a user issue, right? Those didn't necessarily translate directly often to when we were trying to classify the root cause for our data, the data licensing product. Instead, we had to almost look at, okay, where did this data issue come from?

And was it a problem of not having enough source data?

 Chun:

yeah.

 Mariena:

Was it a problem of conflicting data? Things like that. And so we almost had to come up with a whole different taxonomy when we were looking at that root cause. And for sure the customers, like, I mean, even the software ones, they're not going to understand or talk to the root cause of those types of issues, but for data even more so.

And so we, we really just had to come up with a whole new taxonomy. And to do that, we had to go sort of one by one through, you know, a sample. Of of tickets and really work very cross functionally with product and with the engineering team and think about not just I would say the root cause, but what is it that will help us take action?

So, for instance, by tracking and understanding if the issue was the quality of the data we were ingesting versus not having enough data. That could inform the business development team for how to approach our partnerships, right? Cause in this case, we were ingesting third party data. So I think you also have to look at it from that lens of what's going to help you take action and not just sort of classifying things in the way that maybe feels like this is sort of, I don't know, the most accurate.

 Chun:

Yeah. I like it. I like it. The triage system, like on our platform, it came from a design partner, it's from very early days. So then that originally from one of the problem, we find that like, okay, the taxonomy that customers have in their mind are very different than the taxonomy, like organizations have in their mind. And then first time for most of company, like when we were presenting this triage, you're like, oh. Is this kind of like just a labeling system for like customer issues? I'm like, no, like, think of this as more like a routing system internally. You can use this like triage system to like, triage, triage to different product errors or triage to like different root causes to triage to different, like, initiatives. It totally depends on like, what kind of a taxonomy is the best one to, for you to take action.

 Mariena:

Yeah.

 Chun:

I definitely love that. As we both know, it's probably everyone know like for it takes time and takes like resources from engineering and sales and product to really and partnership to really like improve problems like data quality or scale of data.

This whole process, like, how do you, how do you keep the relationship with that customers? How do you make them like, understand this kind of resources constraint there, but at the same time, like, not discourage them to give you more feedback.

 Mariena:

yeah, I think managing expectations is really important in that as well as just keeping an open line of communication. I think with data products, especially and licensing data, I think we've all think about it as partnerships to like, we don't want to just be in a vendor relationship where it's, you know, You know, sort of one sided.

We're giving you this thing, but rather we're helping each other improve. And so we're helping you build a product with this data and you're giving us feedback so that now we can have this cycle of improvement of the data itself. Talking about managing expectations, I think a piece of it comes in where when you're getting that feedback, it's it's understanding the context behind it.

Right? Sometimes you're getting feedback. That is very sort of just backhanded. Like, we see this. We see that. And they're they're just giving it to you because they see it. They don't necessarily even have an urgency or a strong need to see it resolved. And you have to really distinguish that kind of feedback from feedback that is, hey, this is blocking us, you know, really getting value from that data, right?

And having. Building a valuable product. And so that you can actually prioritize that. That's one piece of it. And then once you even know that, though, you have to internally figure out how do you deal with that feedback and what kind of, you know, effort is going to go into You know, doing something with that feedback.

If your organization isn't really set up to react quickly, then customers need to know that if it's going to be, you know, we'll come back in a quarter or we review these things quarterly, whatever the organization is set up to do, and I wouldn't say there's one right answer, but the one thing you do have to do is you have to talk to your customers, manage those expectations so they know either when you're going to review the feedback.

As well as once you've reviewed it, what are you doing with that feedback? Right.

 Chun:

yeah, yeah. That reminds me of the LinkedIn post I posted about last week with all the AI customer support bot you will see this, like, really, really long, just, like, totally useless, like, conversation that are trying to make it a little more personalized, but trying to scale their like customer communication, but from customer side, it's like, it's not telling me anything about like timeline is not giving me any, like, context. So I would say that's probably 1 opportunity, like, for AI to look into it or improve.

 Mariena:

Yeah, yeah. And I think that's been the hard part is doing that. Well, I think you're, you're only really getting that level of relationship management with an enterprise customer success. You're definitely not going to get it from a customer support agent because of the volume of tickets with, and they also maybe don't have, they're not as close.

To the process to get that information, maybe from product and engineering, for instance, so definitely interesting opportunities for AI to make that whole process more efficient because even on the enterprise side, when you have the people doing that, as I discussed, it's a pretty people heavy process and it can be pretty disruptive to the organization.

 Chun:

Yeah, yeah, yeah. I love that. I love that. It's a question. Another interesting question from my own experience, like, onboarding, like enterprise customers. We mentioned about, like, the onboarding effort of like data products and a lot of time is about, like. Data literacy education, and it's really hard to, like, scale that during, like, onboarding, because you're talking with, like, people probably, like, one by one sometimes, right? What are some, like, ideas or practice that you've done before to help, like, customers understand architecture of the data? Or how to use like data for storytelling in a short time or how to read the data to

 Mariena:

well, I mean, in the analytics product that we had, I think the key was to start for at least productizing it from a SAS perspective, start to identify the different types of stories that come up in the data and almost look for the stories for them through templates. And then surface those, right.

And ideally you're moving away from kind of just modules like reports and you are able to, I think, create the stories. But I think there's a lot of opportunities for generative AI to, you know, help with, help with that as well. But I think if kind of extracting some of the different types of things that come up and then sort of looking for, for this customer, is this one applicable or not

And  working with AI to really. You know, showcase that to them. I think, I think it's going to lower the bar, right. You know, in the past sort of larger organizations have been able to do this because they've had the resources to have, you know, the sort of data analysis, data analysts, and maybe business analysts or different combinations of roles that could put that story together and support like a senior leader, for instance, or

Chun:

Yeah.

Mariena:

You know, that product manager, and now it's hopefully going to make a lot of this more accessible.

 Chun:

Yeah. Yeah. Yeah. One of the my favorite features of our chatbot is like, we don't give you a blank bot, like blank paper to start with that. We give you kind of suggested questions based on the data we see in your workspace. And I'm going to say, like, I think most of folks would like, start with, like, suggest other questions first, and then kind of going through this process.

Okay, maybe I should ask, like, a gentleman areas questions there. Awesome. Awesome. Let's see. We talk a lot about the customer onboarding. We talk a lot about the you know, manage expectations. I want to dive a little bit deeper into the process. Okay. That you have helped like organization come up for better kind of like a customer operations there. I know. Well, I would just like you to tell the story. Tell me a little bit more about, like, you know, how you help organizations understand the importance of, like, setting up this, like, really good, like, triage, like a routing of customer operations and how you, like, Yeah. Get like a product and the engineering and the sales team on the same page about, hey, this is what we need to do on like customer management and the customer communication site.

Mariena:

Yeah, I think in terms of support, especially some of the organizations that I've been in have not, when I've kind of entered, they haven't had strong processes, if any, that are well defined for how we deal with issues. And so, you know, these have been more like enterprise SAS organizations and, you know, you might have a customer success manager engaging with.

Someone on a team like I've had, who's sort of their, their technical person to help figure out like what's actually happening and route it and, you know, figure out, okay, no, it needs to go to this engineering team, things like that. But those don't, it doesn't scale at all and it's not measurable. And at the end of the day, we also often haven't actually had even that formalized like support engineering role, for instance, or you're directly going to engineering teams, it gets very chaotic.

And so. Some of the things that I've introduced, you know, some of them are just best practices like having a central ticketing system for internal issues to get raised. And then from there, that's where you can then start to actually really leverage continuous improvement, right? To just. Start to knock things out sort of 1 by 1 really based on whatever is going to be most impactful in your organization.

And I think for me, I'm I'm talking at a minimum at a quarterly level to various stakeholders to both, just get a sense of what is the most impactful. Is it focusing on how we triage? Is it focusing on how we move issues? We can't resolve into the engineering teams. Is it transparency and reporting? So I think that conversation it's good to always.

Keep your nose on that so that you can be adjusting, you know, just like product managers are adjusting roadmaps on the customer support side to like using those same principles of like roadmap planning and sprints, I think has been helpful to create like that culture of continuous improvement.

And instead of coding, sometimes we're working more on process related stuff or. Or maybe we're building small tools. And then some of the, the other things that have been impactful has been like actually creating like routing systems you were talking about. Your, your guys's system has this sort of concept of, of routing through different labels, but you know, we've done that as well.

Introducing SLA is a lot of these organizations didn't have SLAs for support service level agreements, or I guess objectives maybe if they're not. No contractual, I think that goes a long way to getting alignment between the technical product and customer facing teams around when, like, what levels of responsiveness are needed for what levels of issues so that everything doesn't feel like a fire and you're not, you know, sort of.

Really just draining everyone and burning everyone out. So that's me is always the lens. I'm going through with a lot of these process improvements. It's not process for processes sake, but it's really about what processes will help different teams in the organization work better together. So there's less friction.

 Chun:

Yeah.

 Mariena:

We can manage expectations better. And so that the, the people working on these types of issues are actually enjoying their job. They're having fun. They're learning, you know, we're, we're not just kind of burning people out because that can easily happen. And some of these sort of technical support solution roles as well You know, the data, the data rolls.

 Chun:

Yeah. Everyone love like serving customers. I mean, it's also another interesting case to see, like, what is the best way to like phrase customer feedback to motivate the team to work on? We have this like, interesting observation that, our categorization of appreciation. So we can like auto detect, like, what are the things that people are saying to love about your product? And then if we  connect that and send that to like Slack channel engineers, I feel super motivated that when I see the data, I come in.

 Mariena:

It's interesting you bring that up because one of the things that Foursquare did when I was there that I really liked is they would do regular, like pulse surveys so that as people managers, we could understand you know, and. Corrective need be where things weren't working well and what my team always scored the highest sometimes 100%.

I had multiple I think months and quarters where it was 100 percent on the question. Do you understand the value of your work and how it relates to customers or something to that effect? And I always thought it was really cool. Cool. And it, to me, it was also very intuitive, right? We are the ones that were the glue between the product and the customers.

We were the ones really making sure any last mile solutions, you know, got delivered or that they understood things that might keep them from getting value in terms of like that technical education. So I think anything that you can do to help other people who aren't as close to the customer and get that kind of feedback,

 ​It helps them understand how their work is helping people.

I mean, that's why I got into these roles because that's what I wanted. And I wanted to make sure I was working on products that were actually solving problems for customers and get to see that firsthand. And it's a, it's a really fun place to work, but I love that you're finding ways to help other people get that joy as well. So they can also experience

Chun:

I love  that. I love that. I love that. I don't think it's many like organization that are realizing this because we spend so much time like organization spend so much time and money trying to boost the morale there. But by the end of day, like, the magic and power happens. at like the point of we can connect customers with builders and that's it.

 Mariena:

Yeah. I think there's a lot of software engineers, especially that get into that. Career because they enjoy solving problems inherent, right? And so when you can see that something you've done solved a problem for someone that's really satisfying, I mean, obviously we all get motivated by different things, but I think for a lot of people that in of itself, it makes you feel like what you worked on matters.

 Chun:

I love it. I love it. But that half hour went really fast. All right. I don't want to take too much of a time, but I would love to kind of like end on a question that asks everyone what has been one or two like books that you read, read like recently that has been like really impactful and you. Love more people to read it.

 Mariena:

Yeah, well, it's, you know, it's not necessarily technical or related exactly to what we're talking about, but the book I read in the fall that was really impactful for me was Braiding Sweetgrass, and it's written by an indigenous woman who's also a A researcher, she's an ecologist and teaches ecology university level and she, the whole book is about that intersection of science and her indigenous culture and how they approach the relationship with the earth and.

then the scientific view of the earth. And so she talks about data. She talks about the scientific method and experimentation and, you know, how we approach those things with that scientific mind and the sort of other side of that. And it's, it's really cool to see that intersection and open my eyes. So I highly recommend that book to anyone.

Breeding, Breeding Sweetgrass.

 Chun:

Awesome. Awesome. It's also very fitting to the spring theme.

 Mariena:

Yes. And I mean, it does have, you know some context in the world of LLMs, right? We're, we're using a lot more energy as we run large language models. And so a lot of the things she's talking about with our relationship to the Earth and things like that, right? There's, there's a lot there.

Definitely a through line there. So highly recommend it.

Chun: 

I love it. I love it. Well, thank you so much. Is there anything else I haven't asked you and you want to add?

Mariena:

 Yeah, I mean, so my actually most recent venture, I, I switched gears a little bit. So I talked in the beginning about democratizing tech, and that was something that for me has been so important that I actually, on my side, I've started my own business now, Melonhead, and really want to focus on helping individuals as well as different organizations really, you Get the value of all these different innovations in tech that are happening.

So generative AI is what I'm focusing on right now, doing both community focused workshops for individuals to help them really feel more empowered and less from a place of fear with generative AI, and then for organizations as well, helping them actually deploy. Generative AI solutions in their organization.

So hopefully between both of those, you know, we'll see the really impact of generative AI and its potential really trickle down to, you know, all, all kinds of people in society is sort of my goal.

 Chun:

Awesome. Well, again, thank you so much. We, I really enjoy it. I got conversation. I already like learned a lot again from you. And yeah, and thank you everyone for time and we will see you next time.

Summary

Hello everyone. My name is Chun Jiang, CEO and co-founder of Monterey AI. Monterey AI helps companies make it super easy to aggregate triaged and analyze user feedback from sources like support tickets, sales, transcript, and so many more. So the team can always keep a pulse on customers' voices. Make users happy and the unlock, the next growth opportunity. Today, I am super excited to welcome you to our podcast. Making waves. Where we feature leaders in a voice of a customer space to share their stories, practice, and insights.

Today, our guest is Mariena Quintanilla. Formerly a Foursquare and Factual. And the leader in client and sales solutions engineering. Who immerses herself in customer operations daily. And the call herself a data nerd. With her engineering background, we dive into you the fascinating aspects of customer advocacy in the data product world. How do you navigate organizational systems, how to manage customer expectations and AI and accessibility.

If you like our podcast and the content, don't forget to subscribe and follow us on monterey.ai.

Without further ado. That's welcome Mariena.

If you enjoy this podcast, don’t forget to subscribe on YouTube, of follow us on Monterey AI.


Speakers

Where to find Mariena Quintanilla:

• LinkedIn: https://www.linkedin.com/in/mariena/

Where to find Chun Jiang:

• LinkedIn: https://www.linkedin.com/in/chunonline

• Website: https://www.monterey.ai

Transcripts

Chun:

Hello everyone. My name is Chun Jiang, CEO and co-founder of Monterey AI. Monterey AI helps companies make it super easy to aggregate triaged and analyze user feedback from sources like support tickets, sales, transcript, and so many more. So the team can always keep a pulse on customers' voices. Make users happy and the unlock, the next growth opportunity. Today, I am super excited to welcome you to our podcast. Making waves. Where we feature leaders in a voice of a customer space to share their stories, practice, and insights.

Today, our guest is Mariena Quintanilla. Formerly a Foursquare and Factual. And the leader in client and sales solutions engineering. Who immerses herself in customer operations daily. And the call herself a data nerd. With her engineering background, we dive into you the fascinating aspects of customer advocacy in the data product world. How do you navigate organizational systems, how to manage customer expectations and AI and accessibility.

If you like our podcast and the content, don't forget to subscribe and follow us on monterey.ai.

Without further ado. That's welcome Mariena.

Mariena:

Sure, thanks for having me trying. I'm super excited that you guys have started this podcast and I'm excited to talk today. So I, you know, I've been working in tech for about 17 years and have been a little bit all over the place. Bit of a zigzaggy career. I've just been always someone who has.

Really broad interests and likes learning about everything. And I, I knew from early days that I didn't want to just be in software engineering. I wanted to really understand businesses as a whole. And so, you know, I got into product, I got into you know, the customer and like go to market side, building customer facing engineering teams.

Specifically within ad tech and location and pretty early on in my career, I started working with data. And that's really where I, I fell in love and was like, I know I want to work with data and machine learning and it's been so much fun. And I've just really enjoyed helping customers, especially get to use data products and get the value from machine learning and from a that that's what really excites me.

One of the CEOs Gil Alba is actually a four square used to talk about it as democratizing tech. And that really resonated with me. So that's a little bit about me what I care about. I'm here in Los Angeles, originally from outside of Seattle, Pacific Northwest. So always up for hiking or just hanging out with trees and anything really green.

 Chun:

I love it. I love it. Your career trajectory or like experience is super, super interesting. I mean, I'm on kind of the same boat of every time, like, I'm so interested in all the specs of like, business, like, from product design, and then product, and then now customer operations and data analytics in general I would love to hear a little bit more about the original story, like, Walk us through how you find, like, fell in love with a different spaces or different, like, very different roles, right?

In the tech industry and how you find that kind of sweet spot, like, do everything.

 Mariena:

And I mean, and it wasn't, you know, I didn't necessarily get it right away. I think it took trying some things to figure out sort of what I was looking for. So I'll start with sort of the functional side. So I was in software engineering doing application development and after six years of that, you know, wanted to try something else.

And my manager at the time really believed in me and she gave me a shot. She took a risk on me and moved me. From an engineering manager role. I went back into an IC role as a data quality engineer. And that, that was really fun to try something completely different. It was a really hard role because it was a little bit program management trying to influence VPs who are architects, who are product managers.

And I was maybe, I don't know, my early twenties at the time. And so I was surrounded by people with 10 plus years more experience with me, which was such a cool opportunity to learn and be exposed to the way they were thinking and making decisions. And I was mentored essentially by a team from Motorola.

That had come out of you see Chicago I think Urbana Champaign, and they were just true experts on data quality. And they mentored me and sort of trained me on how to think about data quality data quality, best practices and principles. And I became the subject matter expert for Advertising systems at Yahoo, and because I already had 6 years of experience in that business unit, I really understood our advertising systems pretty well.

And then they gave me that sort of functional context of how to now approach it from a data quality lens. And it was. Really exciting to be able to follow the data from ad serving all the way through it getting sort of logged and ingested and processed by data systems and then through billing and reporting.

And I think part of what really made it interesting to me was that I grew up with a family of storytellers, and I started to see that within data, there were just stories all around us, and it was really cool to try to find the stories and the data, and really create that narrative and that, that just, I don't know, it still gets me excited today, you can, if you're on video, anyone watching on video, you can see like, I'm just, I'm excited probably hear it in my voice.

And from the data quality lens, the sort of ability of then good or bad data to really change that narrative was something I saw really early on and I kind of likened it to there could be potholes throughout this sort of road, right? That the data, the journey of the data was taking or even bigger than potholes.

You know, all kinds of road issues that might close down even a road. So I think that helped me see the importance of data quality and the need to really look at it early on, but really throughout that journey. And that's kind of where I fell in love with data.

The second half of your question in terms of functions I wanted to get closer to the customer to be honest.

And so from engineering, I moved into product management and I, I had known and worked with a lot of product managers and I thought that was going to be the Holy grail for me. And four years into product management, I realized I missed being closer to the technology. I wasn't as close to the customer as I thought I would be.

You're really being pulled in so many directions and products and it was incredibly hard role. I really commend anyone in response. So much respect for product managers because it's, it's so hard. You're pulled in so many different directions. And so from there I looked for a role that would be more customer facing and more technical.

I didn't know what that would be. I just started talking to people in my network and kind of talk. Here's what I'm good at. What do you think I should do? What do you know of? And I landed in solutions architecture and this space that is really the intersection of product, of technology, of customers has really been something that I fell in love with.

And so I've spent the last seven years there, both in pre and post sales role, sales engineering support engineering. And it's, it's such a fun time because you really get to see how customers you get to see how customers, the products impact them and you directly solve their problems, which is really exciting.

 Chun:

Love it. I love it. It's still a lot of, lot of. Responsible responsibility to untangle there. It's funny when I was thinking about hiring one sales engineer this morning, I'm like, well, this sales engineer also need to do this kind of customer engineer. This need to do a lot of growth engineering. You need to do a product engineering. And I was like trying to label you to see, okay, what does Marina know? How do we find someone who's like her? And the, one of the biggest challenge, especially like in the field of like AI or this days is everything is getting like even more complicated from my data perspective.  

And what I find, at least like when we talk with customers or onboard customers to Monterey AI We found that most of the time we are not only teaching customers how to use the tool, but also to help them understand, like, what does like data look like in their organization? Where do they see the most available data? Is it in like Salesforce? Is it in like send desk? What are, what is the cadence they're looking at? What are the primary metrics that they're trying to track there? Okay. A lot of this are like data educations  to, to some sense, and that AI education and then tool education,

​ I would love to just learn more about, like, the onboarding secret you have, like, what is the best way to think from, like, customer perspective, but also really smoothly. Like onboarding into this, like, massive of data knowledge and a knowledge to some sense.

 Mariena:

I think you know, that's a, it's a hard question to answer and it's definitely not one size fits all, but I think one of the things I learned when I was in solutions architecture was I think really that value of the discovery phase. And, you know, the, the company I was working for was a marketing analytics platform that was, you know, Taking data in from customers, all the advertising data and then building a machine learning models on top of that to help them understand how their media was performing.

And basically, you know, whether they should buy on Facebook or spend more on TV, things like that. But what what I learned through that experience was really the importance of that discovery phase ideally before you even. Pitch a customer on exactly what you're selling to them, depending on, I guess, the complexity of the offering.

But it was really needing to actually understand. And dive deep into their data. And so that's what I would do is I would leave these discovery calls and kind of came up with a template that would really help me guide those conversations to understand. What do they know? And what do they know? Not know about their own data, right?

Do they have blind spots? Because you're really trying to assess the maturity of the organization because, you know, If it's a immature organization, then that's a really strong signal that they don't have good data hygiene and you're going to run into all kinds of issues. And so, you know, pretty early on that, well, then that just means higher costs, right?

More time, more money. And so you, you kind of need to either bake that in or you need to make the, Decisions in your own organization about are those the customers we want to work with, especially if you're a small startup, right? And you might need to focus on the customers where your product is going to be successful.

So I feel like there, there's definitely one angle there about you know, choosing which customers you're going to work with that have the height for your product until you start to develop maybe more mature. ingestion processes, right? That can themselves maybe do some of the cleaning and deal with the really common issues because no one has perfect data.

So you also still then need to develop different either tooling or start to understand the types of patterns, you know, that your product sort of requires to be able to really find and get rid of that bad data that maybe all customers have certain types of data issues. So that takes time though.

Chun:

It definitely  takes time. Voice of a customers, this kind of concept, like. In other organizations or in non data or analytics products is already very complicated. It takes a lot of like, you know, stakeholders to move the needle there. But I would be curious, like say mission about like after onboarding, right after the customers are like using the platform after they start seeing the data a lot of like voice of customer, like feedback not only really related to the tool they're using or would provide. But mostly, like, this year, this kind of ongoing discovery, ongoing, like, education about their pipe data pipeline, their data hygiene, ​ I would love to hear a little bit more from, like, your experience about that perspective.

 Mariena:

Yeah, I think you know, and the products I've worked on have been kind of a mix of data coming in that isn't from customers that then is being used to create the data products sold to the customers. And in that case, the issues are a bit different from when you're ingesting, say, in that marketing use case, ingesting data, and then giving the customers a product on top of their 1st party data.

Things like that, but I, I think in terms of ongoing education for, for them to clean the data, one of the things that I think has been helpful is really how do you expose some of the metrics to them that can help them be aware of, things that are going to influence their, their ability to really give value from the product, right?

And so it's almost like you know, within engineering, we have, we talk about like health checks and we have different, you know, monitoring and dashboards that we use as engineers and with product managers to sort of assess the health of the product, but different from maybe the metrics you're sharing with executives and senior leaders for, you know, health.

The, the business health, right? But a lot of those are sort of leading metrics or sort of secondary metrics that I think can, can help detect issues before they come up and they better help us understand what's driving maybe those other metrics. Right. And so I, sometimes it's helpful to even have pages within your products that help customers see some of those things.

And, you know, maybe your front page, it can redirect and say, Hey, some of these things we're seeing that, you know, they're going to. It's like the warning warning signs, right? That, you know, if you fix this, you know, you're going to be able to get more value out of, you know, maybe this feature or that feature, things like that.

 Chun:

so it's a lot of, like, if then kind of scenario play there for the customers.

 Mariena:

I mean, potentially, I think it can also be depending on the maybe machine learning base. I don't know if it's heuristics or machine learning, but the types of, you know, you, you tend to know some of the different types of data issues that you're fixing during onboarding. So sometimes it's the setting up checks for those types of things and integrating the same type of data QA that you might do during onboarding.

Into like your ongoing processes and it's going to be phased, right? You might do it one time during onboarding as sort of your first bar and then start to work up to how do we monitor for some of these things so we can send the customer an email or have the account manager reach out when these things go off.

And then eventually longer term, get more towards that self serve Nirvana state of proactive, right? I think they call it like proactive customer support, right? Where you can just reach out right away before. Who affects them?

Chun: 

I love it. What has been, what was the most like challenging? I would love to hear like one story or just like one example about a customer came to you with a problem or maybe with a solution suggestion that is like totally not actually the root cause which happens a lot. I think like in customer's life. Operation Door Success is like, okay, customers, they are super loved, they're friendly, they wanna tell you like how to solve the problem, but sometimes, especially in like highly complicated platform, that is not actually the root cause. I would love to hear a bit more from interacting with like customers. Like how do you get to the next step to find the root causes? For data products.

 Mariena:

Yeah. Yeah. I think for data products, sometimes you really do have to go sort of multiple levels and understand what's the source of that data issue. When we were, you know, originally I had led an effort at Foursquare to classify all of our issues by root cause and the sort of more traditional, I would say software engineering root causes you might think about, was this related to the ability of the tech to, you know, scale.

Was it a bug in the software, you know, things like that, or some infrastructure issue or a user issue, right? Those didn't necessarily translate directly often to when we were trying to classify the root cause for our data, the data licensing product. Instead, we had to almost look at, okay, where did this data issue come from?

And was it a problem of not having enough source data?

 Chun:

yeah.

 Mariena:

Was it a problem of conflicting data? Things like that. And so we almost had to come up with a whole different taxonomy when we were looking at that root cause. And for sure the customers, like, I mean, even the software ones, they're not going to understand or talk to the root cause of those types of issues, but for data even more so.

And so we, we really just had to come up with a whole new taxonomy. And to do that, we had to go sort of one by one through, you know, a sample. Of of tickets and really work very cross functionally with product and with the engineering team and think about not just I would say the root cause, but what is it that will help us take action?

So, for instance, by tracking and understanding if the issue was the quality of the data we were ingesting versus not having enough data. That could inform the business development team for how to approach our partnerships, right? Cause in this case, we were ingesting third party data. So I think you also have to look at it from that lens of what's going to help you take action and not just sort of classifying things in the way that maybe feels like this is sort of, I don't know, the most accurate.

 Chun:

Yeah. I like it. I like it. The triage system, like on our platform, it came from a design partner, it's from very early days. So then that originally from one of the problem, we find that like, okay, the taxonomy that customers have in their mind are very different than the taxonomy, like organizations have in their mind. And then first time for most of company, like when we were presenting this triage, you're like, oh. Is this kind of like just a labeling system for like customer issues? I'm like, no, like, think of this as more like a routing system internally. You can use this like triage system to like, triage, triage to different product errors or triage to like different root causes to triage to different, like, initiatives. It totally depends on like, what kind of a taxonomy is the best one to, for you to take action.

 Mariena:

Yeah.

 Chun:

I definitely love that. As we both know, it's probably everyone know like for it takes time and takes like resources from engineering and sales and product to really and partnership to really like improve problems like data quality or scale of data.

This whole process, like, how do you, how do you keep the relationship with that customers? How do you make them like, understand this kind of resources constraint there, but at the same time, like, not discourage them to give you more feedback.

 Mariena:

yeah, I think managing expectations is really important in that as well as just keeping an open line of communication. I think with data products, especially and licensing data, I think we've all think about it as partnerships to like, we don't want to just be in a vendor relationship where it's, you know, You know, sort of one sided.

We're giving you this thing, but rather we're helping each other improve. And so we're helping you build a product with this data and you're giving us feedback so that now we can have this cycle of improvement of the data itself. Talking about managing expectations, I think a piece of it comes in where when you're getting that feedback, it's it's understanding the context behind it.

Right? Sometimes you're getting feedback. That is very sort of just backhanded. Like, we see this. We see that. And they're they're just giving it to you because they see it. They don't necessarily even have an urgency or a strong need to see it resolved. And you have to really distinguish that kind of feedback from feedback that is, hey, this is blocking us, you know, really getting value from that data, right?

And having. Building a valuable product. And so that you can actually prioritize that. That's one piece of it. And then once you even know that, though, you have to internally figure out how do you deal with that feedback and what kind of, you know, effort is going to go into You know, doing something with that feedback.

If your organization isn't really set up to react quickly, then customers need to know that if it's going to be, you know, we'll come back in a quarter or we review these things quarterly, whatever the organization is set up to do, and I wouldn't say there's one right answer, but the one thing you do have to do is you have to talk to your customers, manage those expectations so they know either when you're going to review the feedback.

As well as once you've reviewed it, what are you doing with that feedback? Right.

 Chun:

yeah, yeah. That reminds me of the LinkedIn post I posted about last week with all the AI customer support bot you will see this, like, really, really long, just, like, totally useless, like, conversation that are trying to make it a little more personalized, but trying to scale their like customer communication, but from customer side, it's like, it's not telling me anything about like timeline is not giving me any, like, context. So I would say that's probably 1 opportunity, like, for AI to look into it or improve.

 Mariena:

Yeah, yeah. And I think that's been the hard part is doing that. Well, I think you're, you're only really getting that level of relationship management with an enterprise customer success. You're definitely not going to get it from a customer support agent because of the volume of tickets with, and they also maybe don't have, they're not as close.

To the process to get that information, maybe from product and engineering, for instance, so definitely interesting opportunities for AI to make that whole process more efficient because even on the enterprise side, when you have the people doing that, as I discussed, it's a pretty people heavy process and it can be pretty disruptive to the organization.

 Chun:

Yeah, yeah, yeah. I love that. I love that. It's a question. Another interesting question from my own experience, like, onboarding, like enterprise customers. We mentioned about, like, the onboarding effort of like data products and a lot of time is about, like. Data literacy education, and it's really hard to, like, scale that during, like, onboarding, because you're talking with, like, people probably, like, one by one sometimes, right? What are some, like, ideas or practice that you've done before to help, like, customers understand architecture of the data? Or how to use like data for storytelling in a short time or how to read the data to

 Mariena:

well, I mean, in the analytics product that we had, I think the key was to start for at least productizing it from a SAS perspective, start to identify the different types of stories that come up in the data and almost look for the stories for them through templates. And then surface those, right.

And ideally you're moving away from kind of just modules like reports and you are able to, I think, create the stories. But I think there's a lot of opportunities for generative AI to, you know, help with, help with that as well. But I think if kind of extracting some of the different types of things that come up and then sort of looking for, for this customer, is this one applicable or not

And  working with AI to really. You know, showcase that to them. I think, I think it's going to lower the bar, right. You know, in the past sort of larger organizations have been able to do this because they've had the resources to have, you know, the sort of data analysis, data analysts, and maybe business analysts or different combinations of roles that could put that story together and support like a senior leader, for instance, or

Chun:

Yeah.

Mariena:

You know, that product manager, and now it's hopefully going to make a lot of this more accessible.

 Chun:

Yeah. Yeah. Yeah. One of the my favorite features of our chatbot is like, we don't give you a blank bot, like blank paper to start with that. We give you kind of suggested questions based on the data we see in your workspace. And I'm going to say, like, I think most of folks would like, start with, like, suggest other questions first, and then kind of going through this process.

Okay, maybe I should ask, like, a gentleman areas questions there. Awesome. Awesome. Let's see. We talk a lot about the customer onboarding. We talk a lot about the you know, manage expectations. I want to dive a little bit deeper into the process. Okay. That you have helped like organization come up for better kind of like a customer operations there. I know. Well, I would just like you to tell the story. Tell me a little bit more about, like, you know, how you help organizations understand the importance of, like, setting up this, like, really good, like, triage, like a routing of customer operations and how you, like, Yeah. Get like a product and the engineering and the sales team on the same page about, hey, this is what we need to do on like customer management and the customer communication site.

Mariena:

Yeah, I think in terms of support, especially some of the organizations that I've been in have not, when I've kind of entered, they haven't had strong processes, if any, that are well defined for how we deal with issues. And so, you know, these have been more like enterprise SAS organizations and, you know, you might have a customer success manager engaging with.

Someone on a team like I've had, who's sort of their, their technical person to help figure out like what's actually happening and route it and, you know, figure out, okay, no, it needs to go to this engineering team, things like that. But those don't, it doesn't scale at all and it's not measurable. And at the end of the day, we also often haven't actually had even that formalized like support engineering role, for instance, or you're directly going to engineering teams, it gets very chaotic.

And so. Some of the things that I've introduced, you know, some of them are just best practices like having a central ticketing system for internal issues to get raised. And then from there, that's where you can then start to actually really leverage continuous improvement, right? To just. Start to knock things out sort of 1 by 1 really based on whatever is going to be most impactful in your organization.

And I think for me, I'm I'm talking at a minimum at a quarterly level to various stakeholders to both, just get a sense of what is the most impactful. Is it focusing on how we triage? Is it focusing on how we move issues? We can't resolve into the engineering teams. Is it transparency and reporting? So I think that conversation it's good to always.

Keep your nose on that so that you can be adjusting, you know, just like product managers are adjusting roadmaps on the customer support side to like using those same principles of like roadmap planning and sprints, I think has been helpful to create like that culture of continuous improvement.

And instead of coding, sometimes we're working more on process related stuff or. Or maybe we're building small tools. And then some of the, the other things that have been impactful has been like actually creating like routing systems you were talking about. Your, your guys's system has this sort of concept of, of routing through different labels, but you know, we've done that as well.

Introducing SLA is a lot of these organizations didn't have SLAs for support service level agreements, or I guess objectives maybe if they're not. No contractual, I think that goes a long way to getting alignment between the technical product and customer facing teams around when, like, what levels of responsiveness are needed for what levels of issues so that everything doesn't feel like a fire and you're not, you know, sort of.

Really just draining everyone and burning everyone out. So that's me is always the lens. I'm going through with a lot of these process improvements. It's not process for processes sake, but it's really about what processes will help different teams in the organization work better together. So there's less friction.

 Chun:

Yeah.

 Mariena:

We can manage expectations better. And so that the, the people working on these types of issues are actually enjoying their job. They're having fun. They're learning, you know, we're, we're not just kind of burning people out because that can easily happen. And some of these sort of technical support solution roles as well You know, the data, the data rolls.

 Chun:

Yeah. Everyone love like serving customers. I mean, it's also another interesting case to see, like, what is the best way to like phrase customer feedback to motivate the team to work on? We have this like, interesting observation that, our categorization of appreciation. So we can like auto detect, like, what are the things that people are saying to love about your product? And then if we  connect that and send that to like Slack channel engineers, I feel super motivated that when I see the data, I come in.

 Mariena:

It's interesting you bring that up because one of the things that Foursquare did when I was there that I really liked is they would do regular, like pulse surveys so that as people managers, we could understand you know, and. Corrective need be where things weren't working well and what my team always scored the highest sometimes 100%.

I had multiple I think months and quarters where it was 100 percent on the question. Do you understand the value of your work and how it relates to customers or something to that effect? And I always thought it was really cool. Cool. And it, to me, it was also very intuitive, right? We are the ones that were the glue between the product and the customers.

We were the ones really making sure any last mile solutions, you know, got delivered or that they understood things that might keep them from getting value in terms of like that technical education. So I think anything that you can do to help other people who aren't as close to the customer and get that kind of feedback,

 ​It helps them understand how their work is helping people.

I mean, that's why I got into these roles because that's what I wanted. And I wanted to make sure I was working on products that were actually solving problems for customers and get to see that firsthand. And it's a, it's a really fun place to work, but I love that you're finding ways to help other people get that joy as well. So they can also experience

Chun:

I love  that. I love that. I love that. I don't think it's many like organization that are realizing this because we spend so much time like organization spend so much time and money trying to boost the morale there. But by the end of day, like, the magic and power happens. at like the point of we can connect customers with builders and that's it.

 Mariena:

Yeah. I think there's a lot of software engineers, especially that get into that. Career because they enjoy solving problems inherent, right? And so when you can see that something you've done solved a problem for someone that's really satisfying, I mean, obviously we all get motivated by different things, but I think for a lot of people that in of itself, it makes you feel like what you worked on matters.

 Chun:

I love it. I love it. But that half hour went really fast. All right. I don't want to take too much of a time, but I would love to kind of like end on a question that asks everyone what has been one or two like books that you read, read like recently that has been like really impactful and you. Love more people to read it.

 Mariena:

Yeah, well, it's, you know, it's not necessarily technical or related exactly to what we're talking about, but the book I read in the fall that was really impactful for me was Braiding Sweetgrass, and it's written by an indigenous woman who's also a A researcher, she's an ecologist and teaches ecology university level and she, the whole book is about that intersection of science and her indigenous culture and how they approach the relationship with the earth and.

then the scientific view of the earth. And so she talks about data. She talks about the scientific method and experimentation and, you know, how we approach those things with that scientific mind and the sort of other side of that. And it's, it's really cool to see that intersection and open my eyes. So I highly recommend that book to anyone.

Breeding, Breeding Sweetgrass.

 Chun:

Awesome. Awesome. It's also very fitting to the spring theme.

 Mariena:

Yes. And I mean, it does have, you know some context in the world of LLMs, right? We're, we're using a lot more energy as we run large language models. And so a lot of the things she's talking about with our relationship to the Earth and things like that, right? There's, there's a lot there.

Definitely a through line there. So highly recommend it.

Chun: 

I love it. I love it. Well, thank you so much. Is there anything else I haven't asked you and you want to add?

Mariena:

 Yeah, I mean, so my actually most recent venture, I, I switched gears a little bit. So I talked in the beginning about democratizing tech, and that was something that for me has been so important that I actually, on my side, I've started my own business now, Melonhead, and really want to focus on helping individuals as well as different organizations really, you Get the value of all these different innovations in tech that are happening.

So generative AI is what I'm focusing on right now, doing both community focused workshops for individuals to help them really feel more empowered and less from a place of fear with generative AI, and then for organizations as well, helping them actually deploy. Generative AI solutions in their organization.

So hopefully between both of those, you know, we'll see the really impact of generative AI and its potential really trickle down to, you know, all, all kinds of people in society is sort of my goal.

 Chun:

Awesome. Well, again, thank you so much. We, I really enjoy it. I got conversation. I already like learned a lot again from you. And yeah, and thank you everyone for time and we will see you next time.