Software engineer: AI agent security and evals
Yusef Syed
I build tools that check what an AI agent actually did, not what it said. First-year Math and CS at the University of Toronto. Looking to join a startup team full-time from January 2027, in San Francisco or New York.
- Scopehavensecurity testing for AI agents (solo project)
- effect-tracersigned ledger of what a Linux process tree did; Rust
- ToolEffectBenchprompt-injection benchmark scored from sandbox state; Inspect
- ToolEffectBench papersole author, Zenodo, 2026
- Checks for Checkers papersole author, Zenodo, 2026
- GitHub14 merged upstream PRs (Microsoft, Apache Arrow, W&B) and the full project list
- Résumé (PDF)one page
- Emailyusefmsyed@gmail.com
- LinkedInprofile