Software, automation & data engineering.
I design and build reliable systems — integrations, automations, data pipelines, and the fixes that get a stalled project shipping again. Work comes with tests and clear, agreed acceptance criteria, so you can see it works.
Start a projectFocused, well-scoped work — delivered with proof it works.
Connecting APIs and services cleanly — payments, AI/LLM, auth & databases, and third-party tools.
Workflow automation that removes manual, repetitive work and gives you back time.
Ingesting, cleaning, normalizing, de-duplicating and validating messy data into something you can trust.
Failing builds, broken test suites, flaky deploys, and apps that almost run.
Selected systems I've designed and built end-to-end — a sense of the depth and rigor I bring to client work.
Aggregates and normalizes data from 19+ live sources — with deduplication, polite rate-limiting, resilient adapters that degrade gracefully, and incremental sync so it only processes what's new.
A ranking/scoring engine that scores options on multiple weighted factors and calibrates against observed results over time, improving as it runs.
Orchestrates work across multiple AI models with automatic fallback. Important output runs through automated dual-model QA — one model produces, a second independently checks and validates it.
Builds in an isolated environment and runs automated release checks that catch failures before handoff.
A real-time web dashboard (HTTP API + frontend) surfacing results, status, and the system's own confidence — built for fast, informed decisions.
My work is backed by 39 automated test suites, a deterministic offline end-to-end test of the whole pipeline, cryptographic build-integrity manifests, per-module version gates, and automated secret-scanning — 90+ modules, ~18,000 lines, with green-before-release checks.
No jargon, clear communication, no surprises.
Tell me what you're trying to build or fix, and what "done" looks like. I'll reply with a clear scope and price.