AI Engineer
I build agent systems, tool governance and engineering evals, with experience in design agents at Kuaishou, enterprise AI platforms and code intelligence. Here are my builds, practices and thoughts.
Focused on tool governance, verifiable evals, and multimodal engineering.
Explore the decisions, evidence and feedback behind each system. Select a node to inspect its inputs and outputs.
Record requirements, targets and tool boundaries so results can be reproduced.
Keep artifacts, execution records and verdicts separate.
Fix specific findings, then verify again with the same task.
Keep inputs, artifacts, evidence and verdicts separate so repairs can be verified against the same task.
Record requirements, target identity and tool boundaries under one acceptance contract.
Public methodology illustration; missing evidence remains unknown.
2026 · AI products, trusted memory and legal technology
CompliPilot: contributed compliance retrieval, citation verification and product engineering, connecting analysis to regulation sources.
Led LabMemory, connecting lab discussions, human confirmation and pre-action auditing through trusted decision memory.
Led product and engineering for TwinBuddy, matching travel companions through digital personas and bilateral agent negotiation.
Main-site intelligent delivery, enterprise AI, and university collaboration.
TRIX Companion connects Web, iOS and desktop with AI agents through a native OpenClaw channel.
TypeScript / SwiftAttrax / CompliPilot: product-image understanding, regulation source grounding, citation verification and compliance reports.
TypeScript / PythonTrusted experimental-decision memory and pre-action audit. Annual Top 40 in the 2026 AI Pioneer Future Talent Competition; competition completed.
Python / TypeScriptInteractive Agent Engineering Workbench.
Contact
Email is the quickest way to reach me. You can also explore my public code on GitHub.
