I turn AI research into shippable products.
Senior Full-Stack & AI Engineer. 4+ years building LLM-powered web apps with Python, FastAPI, and React on Azure. I also lead the team that ships them.
Comfortable overlapping with EU and US East-coast hours · Available for remote engagements and AI consulting.
Recent projects
A cut of the AI platforms, agent systems, and enterprise tools I've led or shipped in the past two years.
BDE — Business Due Diligence Evaluation
Weeks of M&A due-diligence compressed into an 8-pillar score that survives the partner meeting — the LLM evaluates criteria, deterministic code does the math.
LeadLyft
Months of user activity turned into reports that read written-by-coach, not by-model — three AI surfaces sharing one context-bundle layer, 30+ FastAPI endpoints.
AnalyzeEQ
Upload a 3,000-page PDF or a Google Sheet and ask it questions in plain English — hybrid vector + keyword retrieval over pgvector and FAISS, billed on Stripe.
Benjitron-9000
An AI agent that runs an executive's day across Trello, Google Workspace, and OpenPhone — every side-effecting action gated by a human approval before it fires.
VERIQZ
Background checks across 9 Latin American countries — pick the country, enter the right national document, pay with Stripe credits, and get a verified PDF in minutes.
Theseus FinSync AI
Pulls .xlsx data from SharePoint via Microsoft Graph and turns it into financial recommendations — one analysis pipeline, two surfaces (form + chat), no re-uploads.
AI Auto Interviewer
Conducts an entire candidate interview end-to-end — biometric identity check, multilingual question flow, dimensional scoring exported as a recruiter-grade report.
Three things I optimise for
Ship AI that ships.
Models are easy. Reliable, cost-aware, auditable AI products that survive contact with real users are the hard part — that's what I focus on.
Production-first thinking.
A 92%-accuracy notebook model is worse than an 85% model that's reliable, observable, and cheap to run. I optimise for the gap between demo and production — evals, cost controls, latency budgets, and graceful failure.
Azure-native by default.
Most of my production work runs on Azure — OpenAI, Cognitive Search, B2C, App Service, Blob. I know where the sharp edges are.