 Hi, this is your host Apil Bhartya and welcome to another episode of TFI. Let's talk. Today we have with us Allison Hueslid, SVP of Product at Starburst. Allison, it's great to have you on the show. Oh, thank you so much for having me. It's my pleasure to host you today. This is the first time we are talking to Starburst. So I would love to know a bit about the company. How old is the company? What challenges, what problem do you solve for customers? Got it. Yeah. So Starburst has been around for a little over six years now. We were founded in 2017 in Boston and ultimately when we were founded it was with the idea and thought that the data lake could become a much larger center of gravity for organizations data platform approaches. And so what we offer to our customers is a fully featured data lake analytics platform built on top of the open source project Trino. And ultimately the goal is to bring all of the capabilities needed together to discover organize and use data within your data lake or around your data lake without having to do the time consuming tasks of things like migrating data and ultimately enable your teams to focus on building differentiating features rather than managing your infrastructure to leverage your data. Is it like wrong to say that we live in a data-centric or data driven world and if yes we are living in a data-centric world we are creating we are consuming huge amount of data. We are storing data wherever either is comfortable depending on the price but then as you also said egress cost can you know migration can be challenged to be able to extract the right value from the data it has to be the right place as well. So can you talk about the evolution of the places where we keep our data data lakes data warehouse data where lake warehouses I mean the the term technologies are evolving. So we're seeing this really great trend or a large movement trend of people thinking again as you mentioned a lot more about where they start store their data in terms of managing their cost but also with with an Ida performance as well and so we're seeing a lot of a lot more use of data lakes happening in terms of leveraging it for day-to-day data applications so really the data lake at this point's moved from being a place where you just store big data to somewhere where you can actually run those data intensive applications it's really that lake house architecture is really what's enabled that to happen as as an alternative to leveraging a warehouse for that so really with starburst we're really focused on unlocking access to that data within a lake in order to support those data application use cases from a cost performance perspective at whatever way you want to turn those knobs and dials to your organization. And can you also tell me data is a crowded space it's also one of the stickiest space as well. Talk about the I mean that's differentiation is a very kind of loose term but what additional value that you bring to the market as compared to your competitors and who would you consider your competitors. I'd say one of the biggest things that that are in consideration when you look both from a competitive perspective but also a differentiating perspective is how people then manage and govern their data on the lake and that really becomes super critical and so that's a lot of the areas of investment we've had to help better service and automate those processes for our customers. And so some of the things that we've launched recently around our role-based access control and attributes access control over the last year allow you to better manage that access to your data but then also things that we announced recently over the last several weeks around automated data maintenance and automated data tagging and other customers to really continue to service their users in a very safe governed way and without a lot of overhead that you might need with some other solutions. What challenges organizations face when they do deal with you know not just regular data but you have to actually you know extract value from that data and how you folks go in and help them so that they don't have to worry too much about the right talent they should continue to focus on as you also rightly mentioned right business application add value to their application versus spending way too much time on data and related challenges. So one of the things that's really important to us as Starbers is enabling the access to data from really anybody within an organization and to your point that shouldn't be just restricted to a certain type of talent or a certain skill set and so some of the things we've actually been investing in within our product are to leverage generative AI to make the capabilities that we offer and to make the more accessible to a wider range of folks. So we now have a couple of things in the product that that leverage generative AI including taking text and translating that into SQL so you can actually execute queries and get access to the answers you're looking for regardless of your SQL skill sets but it also vice versa helping folks that are inheriting SQL and learning and trying to understand and intuit what your applications or work has already been trying to do understanding what the SQL was doing and bringing that SQL back to text and really sort of bringing that full circle now that will theoretically or ultimately enable a broader range of people within the organization to access that data with varying skill sets. What is the rule of culture within organizations so that these organizations have this data first approach so their team are not working in isolation as we used to have you know in early days the whole silo of data. So I'm just trying to understand from your perspective the imparts of culture for today's organizations so they can make best use of the tools available for them at the same time extract most value from the data and also keep the data where it should be. In terms of a cultural perspective we see sort of topics if you consider the topic of data mesh and how that's really not only about actually the access to the data but also how that data is governed and how that data is managed and part of the ethos there is keeping the ownership of the data with the people who are creating that data. And so we're trying we're enabling capabilities within our products in Starburst in order to allow people to create these curated data sets that represent that ownership and access and actually not just ownership but the efficacy of that data the metadata around it how you should leverage that data. And so basically creating these packaged up artifacts that we're calling data products you can actually bring all of that culture of managing your data appropriately and making it accessible to your broader organization in this one artifact that then anybody can leverage and trust within your organization for leveraging that data in an appropriate way to answer certain types of questions. I want to look at Genity AI from two different perspectives one is of course how do you see Genity AI kind of as either workload or Genity AI helping with dealing with data and then we can also talk about the impact of Genity AI on people HR hiring these are two different questions. On the first question in terms of how do we see generative AI both from helping us help our users as well as sort of its impact on our product we're approaching generative AI at Starburst from two angles one is actually to the first point you mentioned how do we leverage it within our product to enable our users to be more successful and actually more satisfied and happy in their day to day work and that's where things like the text to SQL and vice versa capabilities come into play to help them be more productive and get more value out of the tooling on the flip side just generally we know with generative AI you need to have a strong data foundation with access to all of your data in order to train it properly and make sure you're getting the best content or answers possible out of those systems and so that's where we come into play to actually enable that access across all your data whether it's in your data lake or around your data lake we want to make sure you can then leverage all of that data to get the best efficacy the most accurate answers out of your generative AI or most accurate information out of generative AI and so we're continuing to to look at ways that we can help customers leverage our platform for their generative AI use cases in terms of its impact on people and culture I think the approach that's coming to the fore right now and what's clearly is sort of bubbling to the surface is that it's going to be an and not an or you're going to continue to need people to train the models you're going to need people to ensure its efficacy you're going to need to have people who are validating that they're not hallucinating and so I see this as more of an additive play with generative AI it can help take away certainly some of those repetitive maybe lower value tasks that don't really need a human to do day to day while then enabling people actually to do higher value add tasks and also just making sure that those systems and processes are working in safe effective ways can you also talk a bit about the importance of open source for you folks we believe really heavily in open source in fact all of our products at starburst are built on top of an open source product called trino originally founded at facebook to handle petabytes of data and and effectively that's what's at the core of what we are offering to our customers ultimately one of our key value that we want to provide is this concept of optionality and really allow customers to not be stuck within a vendor locked in space but actually have the freedom of choice of where they store their data how they work with their data and not be locked into any one format and so not only are we heavily big proponents of open source software like trino but also open format such as iceberg and really investing heavily and making sure that we create this really open data lake analytics platform that customers can continue to leverage and evolve with over time because we believe that gives our customers the best optionality and flexibility to evolve with their data needs over time listen thank you so much for taking time out today talk about starburst you know the whole evolution of data lakes where we are storing data thanks for all those insights i'd love to chat with you folks again thank you