Jev and System One Models: Calibration Beats Accuracy
An ML engineer's read on TypeSafe AI's Jev: what a non-autoregressive System One model changes for production classifiers, where it fits, and how I plan to test it.
Writing
Notes from building production ML systems and doing applied research — paper explainers, case studies, and research notes on machine learning, AI for software engineering, and LLM reasoning.
Predicting Pull Request Acceptance the Moment It's Opened
What my IEEE COMPSAC 2026 paper found about predicting PR acceptance and review effort for human and AI-agent pull requests from submission-time signals alone.