Guest: Alison Huselid (LinkedIn)
Company: Starburst (Twitter)
Show: Let’s Talk

Starburst was founded in 2017 in Boston with the idea that the data lake could become a larger center of gravity for organizations. The company offers a fully featured data lake analytics platform which is built on top of open source project Trino. “The platform helps organizations discover, organize, and use data within their data lake and around it without having to spend time migrating data and other time-consuming tasks,” says Alison Huselid, SVP of Product at Starburst.

Data lakes are increasingly being leveraged for day-to-day applications, whereas previously they had just been a place to store big data rather than to run data-intensive applications. Starburst is focused on unlocking access to the data within the lake to support those data application use cases and one of the features they have recently launched is role-based access control to better manage access to the data. Starburst has also announced automated data maintenance and data tagging.

Huselid adds that Starburst is leveraging generative AI to make its capabilities more accessible to a wider range of people within an organization since they believe that access to data should not be restricted to a certain type of skill set or job role. People can execute queries and get access to the answers they are looking for using the generative AI to translate text into SQL. Similarly, it can be used to translate the SQL back into text. The generative AI helps users be more productive and get more value out of the tooling.

On the other hand, Starburst is enabling access to all of an organization’s data, whether within or around their data lake, so that they can leverage it to get the most accurate information out of generative AI. However, even though generative AI can help with some of the repetitive or lower-value tasks humans do, it is clear that people are still needed to ensure its efficacy and validate that the AI is not hallucinating.

How data is managed and governed plays a crucial role in culture within organizations and Starburst’s ethos is to keep the ownership of the data with the people who are creating it. The company’s products enable capabilities that allow people to create curated data sets that represent the ownership and efficacy of that data and the metadata around it. Creating these data products can foster a culture of managing data appropriately within organizations and making it accessible to the broader organization.

Open source lies at the heart of Starburst and all of the company’s products are built on top of Trino, which was originally founded at Facebook to handle petabytes of data. Starburst wants to provide optionality and freedom of choice for where customers store their data and how so that they do not get stuck in a vendor lock-in. The company is focused on creating an open data lake analytics platform that customers can evolve with over time as their data needs change.

This summary was written by Emily Nicholls.