Blog
Thoughts on data, AI, HCI, and research. Some posts are cross-posted on Twitter.
-
On the Consumption of AI-Generated Content at Scale
-
In Defense of AI Evals, for Everyone
-
Writing in the Age of LLMs
-
Data Flywheels for LLM Applications
-
Short Musings on AI Engineering and "Failed AI Projects"
-
Comparing LLMs on "Real-World" Retrieval
-
Mid-PhD Research Talk
-
Study Notes for the Berkeley Databases Prelim Exam
-
Thoughts on ML Engineering After a Year of my PhD
-
Notes on ARIES: A Transaction Recovery Method Supporting Fine-Granularity Locking and Partial Rollbacks Using Write-Ahead Logging
-
The Modern ML Monitoring Mess: Research Challenges (4/4)
-
The Modern ML Monitoring Mess: Failure Modes in Extending Prometheus (3/4)
-
The Modern ML Monitoring Mess: Categorizing Post-Deployment Issues (2/4)
-
The Modern ML Monitoring Mess: Rethinking Streaming Evaluation (1/4)
-
Introducing mltrace
-
Predictive Modeling: A Retrospective
-
Modeling Libraries Don’t Matter
-
Get Rid of AI Saviorism
-
Choosing Between a PhD and Industry for New Computer Science Graduates
-
Reflecting on a Year of Making Machine Learning Actually Useful