I design, ship, and operate AI agents that take over real workflows: in your stack, with evals and guardrails.
30 minutes, no pitch. Bring one workflow and leave with an honest read on whether an agent can own it. Building from zero? Bring the whole idea.
Most teams don't need another AI vendor. They need one senior engineer who takes responsibility for AI actually working, and shows up every week until it does.
That's the role. I plug into your team, fractional by default, and own the AI side of your engineering: what to build, how to build it, and making sure it ships. I'm a full-stack engineer and a founder myself, so the rest of the product is in scope, not handed off.
Audit where AI genuinely fits your product and operations, and where it doesn't
Prioritize by impact and feasibility, not hype
Kill dead-end ideas before they burn a quarter
Model selection, RAG vs. fine-tuning, build vs. buy calls
Agent design: tools, memory, orchestration, human-in-the-loop
Cost and latency budgets set before code is written, not after the invoice
I write the code: agents, APIs, frontends, and infra, from prototype through production
Working software every week, not slide decks
Your repos, your cloud, your standards
Eval harnesses and agent patterns your engineers can extend
Code review and pairing on AI features
The goal is a team that ships AI without me, not a permanent dependency
Worth saying plainly: I work solo and keep the roster short. The person on the intro call is the person writing the code.
The flagship work: agents that take real actions in real workflows, backed by the evals, guardrails, and monitoring that separate a demo from a system you can trust.
Read inbound tickets and emails, classify intent, draft responses from your docs and past resolutions, resolve the routine ones, and escalate the rest to a human with full context attached.
Intake, extract, validate, and route: invoices, contracts, forms, and applications turned into structured data in your systems of record, with a human approval step where the stakes demand it.
Permissions-aware retrieval over your wikis, docs, and past tickets, with sources cited on every answer, so "ask the bot" beats "ask around and wait."
Multi-step processes across your CRM, email, and ticketing via tool calls: drafting, updating, scheduling, and reconciling, with approval gates on anything irreversible.
Enrichment, deduplication, recurring reports, and monitoring that watches your data and speaks up when something looks wrong, instead of a dashboard nobody opens.
What "production" means here: an eval suite built from your real cases, guardrails and fallbacks for when the model is wrong, tracing on every run, cost monitoring, and documentation your team can operate from. If it can't be measured and trusted, it doesn't ship.
Workflow not on the list? That's what the intro call is for. And if an agent is the wrong tool, I'll say so on the call, for free.
No big-bang AI projects. One workflow at a time, proven on your real data before it touches production.
Pick one workflow with clear value
Define what "correct" looks like in numbers, plus a cost and latency budget
Audit the data and tools the agent needs
Deliverable: a written build plan with architecture and cost estimate
A working agent on your real data in weeks, not quarters
Ugly on purpose: the point is proving the core loop works
Deliverable: a demo you can put in front of the people who'd use it
An eval suite built from your real cases and hard edge cases, run on every change
Measure accuracy, cost per run, and failure modes
An honest go/no-go, including "don't build this" when the numbers say so
Deliverable: an eval report you can make a decision on
Guardrails, fallbacks, and human approval gates
Staged rollout: shadow mode first, then supervised, then autonomous where the numbers justify it
Tracing, alerting, and cost controls, deployed in your cloud, in your repo
Deliverable: a live agent your team owns
Review transcripts, catch drift, tune prompts and models as they change
Runbook and handoff docs so operating it doesn't require me
Deliverable: a system that keeps working after I step back
A short, fixed engagement: audit your product, data, and team; identify the highest-leverage agent opportunities; leave with a prioritized roadmap and architecture recommendations. Useful on its own, and the natural first step to everything below.
The core offer. A set number of days per week or month, ongoing: I own AI delivery, build agents through the process above, make the architecture calls, and level up your engineers as we go. Senior direction without the recruiting cycle or the agency invoice.
A single scoped agent, end to end: scope, prototype, evaluate, production, and operate, with a defined handoff at the finish.
For zero-to-one teams. I join as your first engineer and build the whole product, agents at the core: from empty repo to MVP, then toward product-market fit. Full stack, fractional or full-time, shaped to the stage.
You have real workflows an agent could take over, but no senior AI engineer to lead the work
You're a founder who needs agents shipped, not another strategy deck
Your engineers are strong but new to LLM systems, and you want the capability in-house when I'm gone
You're pre-product and need a founding engineer who can take the whole build from zero to MVP, agents included
It's not for you if you want someone to validate a decision you've already made, or AI on the roadmap for the optics rather than the work. If the fit isn't there, I'll say so early and point you somewhere better.
The next step is a 30-minute call: bring one workflow or a whole product idea, and you get a straight answer on whether and how I'd build it.