Production application
Microsoft Teams AI assistant
An enterprise workflow reduced from hours to minutes; outcomes and integration constraints.
Read case study →Home / Selected work
Projects and case studies
Start with the production Teams assistant, then see how I adapted developer tools to enterprise restrictions and tested whether extra agent components were useful.
Start here
Developer tooling
Codebase Memory CLI
A CLI-first fork designed around enterprise MCP server restrictions, with upstream attribution.
Read case study →Reliability and engineering judgment
Agent-Workflow
Tools for resuming agent runs and recording results, plus a review of which components are actually needed.
Read case study →Additional applications and studies
Agent-Workflow Comparative Evaluation
When a model-based router picks differently from rule-based routing, that does not show it is better. This project checks which decision was right, using an answer key written by independent labelers who never saw either system's output. The results are published whether or not the model-based system wins.
Filing Harbor
Tax preparation needs arithmetic you can audit and a clear answer to who holds the data. A model can help find or explain a rule, but the arithmetic belongs in fixed code. A public demo should not keep tax drafts that contain names and addresses just because storing them is convenient.
Open Filing Harbor ↗Grounded
An LLM is most useful here for working out what the user means. Application code should keep control of the database.
Open Grounded ↗Semantic Site Evaluator
Tests that match exact wording break when a site's copy and structure change on purpose. Handing the judgment to an LLM would put release decisions in a system that can be wrong. This project records what must stay true instead of today's exact wording, collects limited browser evidence, checks each rule against the evidence it needs, and leaves release decisions to ordinary code.
Rocket Tooling Designer
I wanted the tooling calculations to be usable right away in a browser, and to produce files a manufacturer can use.
Open live demo ↗Agent-Workflow TypeSafe AI
Some engineering decisions depend on meaning, so fixed rules handle them badly. A model can still advise without running the workflow. This adapter turns a few narrow judgments into saved, checkable advice. Agent-Workflow's rule-based code keeps control of policy, execution, review and acceptance.
SpecGen
AI agents are easier to control when requirements, decisions, acceptance criteria, evidence and open questions are written down in files.