Four disciplines, each grounded in systems I've shipped to production and operated over time.
Production RAG, streaming replies, multi-tenant infrastructure.
Hybrid retrieval combining pgvector, BM25, and RRF with a reranker. Vercel AI SDK for Claude and OpenAI orchestration. Operator tools for human-in-the-loop. Two years in production for a European manufacturer, handling 100K+ messages with 80% reply acceptance.
Full-stack TypeScript, built for production.
React, Next.js, and Astro. Type-safe end to end, with Zod schemas propagating from database through API to UI. SSR, ISR, or edge-rendered as required. Payment integrations, authentication, and migrations handled as first-class concerns.
Manifest V3 at scale, without losing user data.
55,000 daily users on a production extension. 26 storage migrations executed without data loss. Service-worker architecture, cross-context state management, offscreen audio, staged rollouts, and Chrome Web Store listing fan-out from a single codebase.
Infrastructure engineered for reliability.
.NET and Azure Functions experience at TCS-scale. Self-managed Podman and Caddy on bare VPS for stateful services. PowerShell and Bash automation. Zero-downtime deployments with health checks and clear rollback paths.
Every engagement starts with the problem, not the framework. Time invested upfront on scope and constraints prevents months of rework later.
Small, reversible pull requests. CI/CD, health checks, and migrations built in from day one — never retrofitted at the end.
I build, deploy, and support the systems I design. Handoffs are documented, predictable, and engineered to be handed off.
Architecture decisions, trade-offs, and postmortems are documented. The next engineer should be able to read the system without talking to me first.
One or two engagements at a time. AI platforms, Chrome extensions, full-stack web applications, and cloud infrastructure. Scoped on the first call.
