Enhance, Don’t Replace: A Recipe for Success in Data Tooling

Abstract: Most data analysis and data science is performed in human-centered tools, such as spreadsheets, visual analytics tools, and data science libraries. However, these tools often pose challenges for end-users, especially those without extensive programming expertise, in terms of scalability, interactivity, and usability.

Rather than forcing such users to switch tools, over the past decade, we’ve instead taken the approach of enhancing existing tools with an eye towards addressing these challenges. To do so, we draw on techniques from data management and human-computer interaction. In my talk, I’ll describe a couple of successful examples of tools that have been adopted widely by end-users. Finally, we also reflect on how our recipe — of enhancing existing tools as opposed to replacing them — may need revisiting in the exciting arena of LLM-powered data work, which forms the focus of our new EPIC Data lab at Berkeley.

Bio: Aditya Parameswaran is an Associate Professor in EECS at UC Berkeley. He works in the broad area of human-centered data science, developing usable, robust, scalable, and intuitive data science tools. His open-source tools have recieved thousands of GitHub stars, and have been downloaded millions of times across a wide spectrum of industries. Ponder, a company that Aditya cofounded with his students in 2021 based on open-source tooling developed as part of research, was acquired by Snowflake, the leading cloud data warehouse vendor in 2023. Aditya has received the Alfred P. Sloan Research Fellowship, VLDB Early Career Research Contributions Award, the ARO Young Investigator Program Award, the NSF CAREER Award, the TCDE Rising Star Award, a number of best paper awards, along with other recognitions. His website is at http://adityagp.net.