.png){kind=logo}
Software Engineer Creates Tool to Track User Behavior in Data Analysis Software
Recent data shows that using outdated data visualization systems and other software solutions can lead to significant financial losses for businesses. In an article The Guardian, the so-called "technical debt" – the cumulative cost of supporting and operating outdated or inefficient technologies – costs the US economy approximately $2.41 trillion annually. These expenses include cybersecurity issues, operational failures, failed development projects, and maintenance of legacy systems.
What if you could replay every decision a data analyst makes, step by step, like rewinding and fast-forwarding through a movie? Imagine catching the exact moment when someone discovered a critical market insight, and then sharing that precise sequence of clicks and filters with your entire team. This isn't science fiction – it's what Kiran Gadhave made possible.
When cancer researchers at the Aardvark project needed to ensure their data exploration workflows were reproducible and transparent, they turned to Gadhave's solution. His innovation has become the backbone of visualization tools that demand meticulous tracking of user decisions.
Gadhave created Trrack, a unique specialized JavaScript/TypeScript library, while pursuing his PhD at the University of Utah under Prof. Alexander Lex's supervision. The library, which has been downloaded over 20,000 times, revolutionizes how developers build interactive tools by creating what experts call a "non-linear provenance graph" – essentially a complete map of every action, filter, and decision a user makes while exploring visual data. He now works as a Software Engineer at Imply Data.
Gadhave's background doesn't fit the typical computer science mold. After finishing his Mechanical Engineering degree at the University of Mumbai in 2015, he worked at an Indian startup called Cadsol, blending mechanical and software engineering skills. In 2017, he made a bold move to the US for a Master's program in Computer Science at the University of Utah. This transition wasn't easy.
"When I started grad school, I didn't have any academic background in Computer Science," Gadhave recalls. "I had software development experience from India, but I'd never taken formal computer science courses or done research. I had to work incredibly hard to catch up with the undergraduate coursework while taking graduate classes."
His breakthrough came when he started working as a teaching assistant for Alexander Lex in an Introduction to Data Analysis course. Impressed by Gadhave's work ethic, Lex offered him a summer job as a software developer at the Visualization Design Lab. By summer's end, Gadhave had earned a spot in the PhD program under Lex's guidance.
At the University of Utah, Gadhave became a research assistant and later a PhD candidate specializing in data visualization. His work includes developing the Trrack library, publishing multiple research papers, and mentoring undergraduate students in data analysis and Python programming. He also conducted training seminars for researchers from non-computer science backgrounds, helping them integrate computational tools into their work.
Now at Imply Data, Gadhave works as a Software Engineer II, where he applies his expertise in data visualization and user interaction tracking to develop enterprise-level analytics software. The company builds real-time analytics platforms used by organizations for processing large-scale data streams.
Gadhave's breakthrough creation is Trrack – a tool that creates a map of every click, decision, and change users make while exploring data visualizations.
Modern data analysis rarely follows a straight line. Analysts often explore multiple branches of inquiry, applying different filters and parameters as they search for insights. Without proper tracking, these complex exploration paths are lost when the session ends - a serious problem for scientific research and decision auditing.
Trrack solves this by creating what Gadhave calls a "non-linear provenance graph" – a structure that captures not just a linear history of actions, but entire branching decision trees. The library saves application states at key moments, allowing users to jump between different points in their analysis or revisit alternative paths they previously abandoned.
Gadhave's 2020 paper "Trrack: A Library for Provenance-Tracking in Web-Based Visualizations" has been cited 67 times, and the software itself has been downloaded over 20,000 times from the NPM registry (the standard package repository for JavaScript libraries). Research groups at the University of Utah and Johannes Kepler University have integrated it into their visualization work.
Gadhave’s development addresses key challenges faced by visualization developers by enabling the precise reproduction of analytical workflows through Trrack, a capability vital for projects like the cancer research initiative Aardvark and clinical decision-making tools such as Sanguine. The library also enhances team collaboration by allowing users to export and import interaction states, streamlining the sharing of insights across teams. Additionally, Trrack improves user experience with features like undo/redo and session recovery, making complex data exploration tools more intuitive and accessible.
The widespread adoption of Trrack demonstrates its real impact across the professional community. With over 20,000 downloads from the NPM registry and active implementation in research institutions like the University of Utah's Visualization Design Lab and Johannes Kepler University's Visual Data Science Lab, the library has proven its value to developers worldwide. This adoption extends beyond academia – data visualization professionals in various industries have integrated Trrack into commercial applications, recognizing its ability to enhance transparency and accountability in data exploration processes.
As visualization tools become more central to business decision-making (the market is expected to reach $19.2 billion by 2027), Trrack's approach to tracking user interactions positions it at the cutting edge of industry needs.
Gadhave has published seven papers and posters at major conferences including EuroVis and IEEE VIS, the flagship conference of the Institute of Electrical and Electronics Engineers, the world's largest association of technical professionals. These publications have been cited 134 times combined.
His work earned presentation slots at EuroVis 2022 in Rome ("Reusing Interactive Analysis Workflows") and the competitive IEEE VIS 2022 Doctoral Colloquium in Oklahoma City, where he presented his PhD thesis "Literate and Reusable Visual Analysis." In 2023, he received an invitation to the prestigious Dagstuhl seminar.
Gadhave's expertise has made him a sought-after peer reviewer for CHI (acceptance rates below 25%), PacificVis, and the Journal of Open Source Software. He also serves as a judge for the Globee Awards for Technology, evaluating innovations in data analytics and visualization tools.
After completing his PhD in May 2024, Gadhave joined Imply Data as a Software Engineer. Earlier this year, he was nominated to join Sigma Xi, one of the world's oldest scientific societies that includes over 200 Nobel Prize winners.
As we move through 2025, the practical impact of work like Gadhave's becomes increasingly clear. His contribution to tracking data exploration history provides a foundation for more transparent and reproducible visual analysis.
The shift from simple charts to rich, interactive visualization tools isn't just a technical improvement – it fundamentally changes how we make sense of complex information. By capturing the full context of how people interact with data, Trrack and similar tools make it possible to understand not just what insights were found, but how they were discovered.
For companies struggling with ever-more-complex datasets, these advancements offer a path to more effective data exploration. As visualization continues evolving, the principles Gadhave has helped establish will likely become standard practice in how we make sense of our increasingly data-rich world.
