Work, 5 spans
Shipped to production,
and built to learn.
Three systems from my jobs, written up without the confidential numbers, and two lab builds that push the same ideas further.
Shipped
- A second opinion for every AI decisionAI validation of advertising creatives, plus a Claude agent that reviews the AI's verdicts so human reviewers know when to look twice.agent.second_opinion
- From booking to live, without the spreadsheetOrchestrating ad creatives across a CRM, two ad servers and AI services, deployed with infrastructure as code and passwordless CI/CD.software.engineer
- Keeping a global corporate site steadyDesign, build and regression testing on avanade.com, with integrations across search, video and marketing platforms, released monthly.senior.software.analyst
Lab builds
- An agent that reads the logs firstAn MCP server that lets Claude query Application Insights and Log Analytics, propose a root cause for an incident, and show the queries behind every claim.lab.root_cause
- Runbooks that answer back, with a scorecardRetrieval over runbooks and architecture decision records, graded by golden questions and an LLM judge on every pull request.lab.runbook_rag