A LangGraph + Claude Agent SDK pipeline that reviews pull requests against security, licensing, and data-handling policy before merge — cost-gated, human-in-the-loop.
- active
- LangGraph
- Claude Agent SDK
- FastAPI
- Policy compliance
Computer-vision researcher (PhD & postdoc, 2018–2023) turned builder of LLM systems that turn regulation and review into structured, auditable output — deterministic decision layers, human-in-the-loop, evaluated.
A LangGraph + Claude Agent SDK pipeline that reviews pull requests against security, licensing, and data-handling policy before merge — cost-gated, human-in-the-loop.
A pip-installable PDPA compliance layer for Malaysian SMEs using AI/LLM tooling with customer data. One shared codebase, two packages: presidio-audit and presidio-malaysia.
Multi-agent supervision of a live ML model: monitoring, investigation, remediation, post-mortem.
vcrpy for MCP. Records real agent–server sessions into diffable cassettes and replays them as deterministic mock servers.
Skills are context in a plain text file. Why Claude choosing when to load them matters, its downside, and why I build skills around workflows, not roles.
Every fix to my incident loop after release was found and verified by reading code. When the demo finally ran end to end, it worked, and that changed what I think it's for.