[● PILOT]
Built at Quantum Strides · sole engineer
Agentic Walkthrough Assistant for Enterprise IWMS
An agent that autonomously learns an enterprise asset-management system by driving its live UI — safely — and turns it into a queryable knowledge graph.
// overview
A multi-agent system that onboards itself to a complex enterprise IWMS the way a new analyst would: by clicking through the real interface. Browser automation drives the live UI while Claude reasons over what it sees, mapping pages, forms, and workflows into a Neo4j + Pinecone graph for retrieval — with a hard guarantee it can never mutate live data.
// the problem
Every IWMS deployment is configured differently, with hundreds of forms and undocumented workflows. Manually mapping one took analysts weeks, and any automated crawler risked accidentally submitting or deleting live records. The agent had to learn fast and be provably read-only.
// architecture
01 · drive
Browser agent
Playwright
02 · perceive
Vision + DOM
Claude reasons
03 · extract
Form schema
fields + sub-grids
04 · graph
Neo4j + Pinecone
GraphRAG
05 · serve
LangGraph chat
walkthroughs
// side rail
Never-mutate guard
code-enforced
⛔ every write physically blocked — zero mutations, ever
// what i built
- ✓Browser-automation layer (Playwright + browser-use) that drives the live IWMS while Claude interprets each screen.
- ✓Vision-based form extraction that opens modals and captures fields, types, required flags, tabs, and sub-grids — including required lookups.
- ✓A code-enforced never-mutate guard (not prompt-based) that physically blocks any Submit/Save/Delete before it can fire — safety is a property of the code path, not the prompt.
- ✓Single-producer / multi-consumer concurrency to respect the IWMS one-session constraint while parallelizing the LLM step.
- ✓GraphRAG store in Neo4j (structure) + Pinecone (vectors), and a LangGraph chat pipeline with a guarded live-walkthrough mode.
- ✓Token-accurate per-run cost instrumentation; shipped under TDD with 900+ passing tests.
// impact
0+
tests passing
0
mutations, ever
~2.7×
faster form extraction
// tech stack
PythonLangGraphClaude (Bedrock)Neo4jPineconePlaywrightFastAPIbrowser-useReact
next project
Multi-Tenant White-Label AI Chatbot