Intelligent systems
- LLM systems
- RAG & retrieval
- Agents & tools
- Evals & observability
Open to selective AI & platform work · Q2 2026
Intelligent systems that scale.AI engineering for production — LLMs, retrieval & agents on real platforms.
I build intelligent systems that ship — LLMs, retrieval, agents, and tooling — grounded in real software: platforms, APIs, data paths, and UIs. Production-grade evals, clear failure modes, and ownership your team can sustain. No science fair; no surprise outages.
Remote-first · Overlap with US & EU
LLMs & retrieval·Agents & tools·Platforms & APIs
Stack
Technology I work with
Production-minded tooling across models, languages, data, and cloud — grouped how I think about shipping systems.
Languages & frameworks
- TypeScript
- Python
- Next.js
- Node
Data & infrastructure
- PostgreSQL
- Kubernetes
- AWS / GCP
Platform & leadership
- Platform & APIs
- Technical leadership
Intelligent systems — the focus
AI that lives inside software: models, retrieval, and agents — engineered with the same rigor as the rest of your stack.
Intelligent systems in production
LLMs, retrieval, agents, and tools — with evals, grounding, latency budgets, and governance so the model layer doesn’t outrun the software you ship around it.
Platforms underneath the intelligence
Services, APIs, data paths, and tenancy so intelligent features plug into something operable — observability, on-call, and clear ownership.
End-to-end delivery
One thread from problem framing to what runs in prod — front end to storage — with trade-offs your exec team can stand behind.
At a glance
10+
Years shipping software
Prod
Intelligent systems, not demos
Global
Remote teams & hybrid
Partner
Embedded with your org
Enterprise SaaSHigh-growth startupsFortune 500 tech
About me
Building calm, reliable intelligent products.
I'm an AI engineer building intelligent systems that survive contact with production — LLMs, retrieval, and agents, on top of solid platforms and APIs.
I care about latency, evals, and clear failure modes as much as demos. If that matches how your team works, we'll get along.
How I work
Intelligent systems need boring reliability.
The best intelligent products feel boring in production: predictable latency, grounded outputs where it matters, and teams that can iterate without fear. I partner with product and engineering on LLM-backed experiences — and the platforms, APIs, and data paths that keep them honest — from framing to metrics that match the business.
Outcome-obsessed
Technical choices tied to user impact, cost, and time-to-ship — not novelty for its own sake.
Calm execution
Clear communication, tight feedback loops, and documentation that survives handoffs.
Expertise
Intelligent systems & the stack beneath them
From LLM product surfaces to platforms and leadership — open an area for how I work with teams in practice.
Selected work
Proof you can trace to production
Representative engagements — anonymized where required — including intelligent products and the platforms they sit on. Context, constraints, and what shipped.
What leaders say
Trust earned in production — not pitch decks
Anonymized where confidentiality matters — from AI rollouts to platform bets. Outcomes I optimize for.
“He gave us a north star for our intelligent product — architecture in plain language, and a path from models to something we could actually run and own.”
“Rare: deep taste in LLMs and the patience to keep governance real. Our assistant shipped on time without cutting corners we’d regret in audit.”
“I trust the trade-offs he documents. Postmortems improved; on-call got calmer. That’s worth more than any demo.”
Engagement model
A cadence built for clarity and speed
No black boxes — you always know where we are, what we're proving next, and how we measure success.
Align on outcomes
We lock the user promise, risk profile, and what “good” means — for the product and for any intelligent layer — before architecture eats the calendar.
Ship thin slices
Vertical slices so model, retrieval, and API risk surface early — with observability baked in from day one.
Harden & transfer
Runbooks, evals, and handoff so your team owns the intelligent system — not a notebook with my name on it.
Iterate with signal
Dashboards and reviews tied to business impact — for core product metrics and for model quality, latency, and cost where AI is in the path.
Track record
Scope that compounds
2022 — Present
Lead Engineer — AI & intelligent products
Confidential · High-growth product org
Own intelligent product surfaces — assistants, retrieval, and model-backed workflows — on top of resilient platforms: services, data, APIs, and the reliability story that keeps them alive in production.
2018 — 2022
Senior Software Engineer
Series B — Enterprise SaaS
Shipped platform-critical features, raised the bar on reliability, and mentored engineers through complex migrations.
2015 — 2018
Software Engineer
Product studio
Built customer-facing apps and APIs across web and mobile; learned to ship fast without mortgaging the future.
FAQ
Questions, answered before the call
Clear expectations lead to better conversations — and faster decisions.
Lead roles building intelligent systems — LLM apps, retrieval and agents, evaluation harnesses, and production deployment — always anchored in solid platforms: services, data, APIs, and UIs. Advisory and exec alignment when you need a credible technical voice.
Writing
Notes on intelligent systems
LLMs, platforms, and judgment — written for people who ship to production.
Building intelligent systems customers can trust?
If you need AI that holds up in production — not slideware — and a lead who can tie models to platforms, reliability, and your roadmap, let’s talk while you still have room to maneuver.
Book a conversationContact
Tell me what you're building — I'll reply personally.
Building an assistant, retrieval stack, or agentic workflow — or threading AI into an existing platform? I help teams ship intelligent systems with clear trade-offs and a path to production.
Send a message