AI Tells Check
by Lawrence Hua
A pre-flight linter that flags common AI-generated "tells" and slop patterns in frontend and marketing code.
Hi, I'm Lawrence — a Senior AI Product Manager and founder who builds AI products that actually ship. I've been building with AI tools for three years, very intensely over the past two years, building over 50 AI products. I love turning messy, real-world work into clear, reliable systems, with evaluation gates and human review before anything goes live, so teams can trust what their agents do. I've shipped AI products at Asurion, led Smart Upsells and kiosk onboarding at GRUBBRR, founded Expired Solutions to cut grocery waste with CV and GPT freshness scoring, and built RAG tools that gave teams hours back each week. My skills focus on fail-closed safety, evidence-gated delegation, and reversible cleanup — practical methods, not hype, made for people who want agents that are helpful, honest, and safe to run in production.
by Lawrence Hua
A pre-flight linter that flags common AI-generated "tells" and slop patterns in frontend and marketing code.
by Lawrence Hua
Retire a dead project without breaking anything. Your agent finds every surface tied to it (repos, scheduled jobs, dashboards, notes), shows you the list, waits for your yes, then archives reversibly with a JSON undo manifest. Hard-delete is banned. Every step can be replayed backwards.
by Lawrence Hua
A framework for letting sub-agents, worker fleets or pipelines act for you safely. Gate each task family on a track record, verify every result with deterministic evidence instead of the delegate's own "it passed", and climb a Shadow → Propose → Autonomous ladder one rung at a time.
by Lawrence Hua
A read-only, fail-closed check your agent runs before deleting any worktree, clone, cache or build folder. It screens scheduled jobs, crontab, config files, running processes and shared git object stores for references to the path, and returns SAFE only when every screen finishes with zero hits.
by Lawrence Hua
Teaches your coding agent to find guards that silently allow everything. When a gate reads a field that might be missing, the default must REFUSE, not allow, and every write-site must actually set the field. Includes a 6-step review checklist and before/after fixes.