news

Aug 17, 2026 Our work on scaling reranker inference, “Drowning in Documents”, was featured on the Weaviate Podcast! Thanks to Connor Shorten for the great discussion on rerankers and where retrieval is headed — check out the episode and the announcement thread!
Jun 26, 2026 Excited to announce the release of TraceLab, a trace of ~4,300 real coding-agent sessions (~350K LLM steps, ~430K tool calls) from day-to-day use of Claude Code and Codex, along with an analysis of what these workloads mean for LLM serving! Check out the website and the announcement thread from UW SyFI!
Apr 12, 2026 Honored to have received the NSF Graduate Research Fellowship!
Jun 19, 2025 Happy to share that our work on scaling reranker inference in RAG pipelines will be presented at ReNeuIR’25 at SIGIR 2025 in Padova, Italy! Huge thanks to the reviewers, my collaborators, and Databricks! You can see the updated paper here.
Nov 27, 2024 My preprint with the Parallel Programming Laboratory at UIUC, “CkIO: Parallel File Input for Over-Decomposed Task-Based Systems” is on the arxiv! If you are using Chrome, make sure to download the preprint and view it on local viewer - some figures come out weird on Google Chrome it seems.
Nov 18, 2024 Our preprint from my Databricks internship “Drowning in Documents: Consequences of Scaling Reranker Inference” is on the arxiv!
May 20, 2024 Started internship at Databricks Mosaic Research as a Research Scientist!