Senior SDET · Agentic AI
Eight years automating tests for FINRA, S&P Global, and Duke Energy. Now building LLM agents, RAG pipelines, and MCP tooling that speed up test automation, and testing them to the same standard.
$ npx playwright test --reporter=json ✓ 209 passed ✗ 3 failed (4m 12s) $ triage --run latest # MCP → LLM → list_failures 3 results → get_failure_detail checkout.spec.ts:48 → compare_history last 10 runs product bug checkout total ignores coupon flaky search.spec timeout on /api/suggest flaky login.spec same failure 3/10 runs schema ✓ evidence cited ✓ report posted
An MCP server that hands Playwright test results to an LLM, so failed runs come back sorted into likely root causes instead of a wall of stack traces.
Tech Horizon private repo
A test harness that runs repeatable conversations between two LLM characters, checks every transcript automatically, and uses an LLM judge for the parts rules can't catch.
Personal project private repo
An agent that turns raw test results and defect data into a daily engineering report a lead can read in two minutes.
Tech Horizon private repo
LLM agents with real tools: triaging failed runs, drafting test cases, writing daily reports. Python and TypeScript, structured outputs, MCP servers.
Agents are only useful where people already work: CI runs, pull requests, Jira, qTest. I wire them into the workflow instead of a separate demo.
Evals, guardrails, schema checks, and trace review. The same discipline I’ve applied to regulated financial systems, now pointed at LLM output.