Production AI Engagements
Production AI for teams shipping into the real world.
I work with engineering teams on retainer to ship reliable AI features, harden the AI stack, and make agentic systems that survive past the demo.
15+ years building production software. Former CTO at a regulated medical-device shop. Current senior engineer at a clinical software company. The patterns I bring are tested with real users, real token budgets, and real error rates — not adapted from blog posts.
How It Works
Every engagement starts with a 2-week paid Foundation Sprint that produces shippable artifacts. After the Sprint, you can extend into a retainer at the intensity you need, or walk away with what we built. No long-term commitment, no calendar-bound deliverables, no surprises.
Foundation Sprint
2 weeks · $12,000 flat
We pick the highest-leverage problem in your AI stack and ship a foundation piece against it. Either side can walk away after, or convert to retainer.
Production AI Retainer
Monthly · 30-day notice either side
Ongoing engagement at the intensity you need: Advisory, Embedded, or Full. Flat monthly rate, flexible hours within the stated range, no timesheets.
Foundation Sprint
$12,000 · 2 weeks
Two weeks to identify the highest-leverage problem in your AI stack and ship a foundation piece against it.
Most engagements that fail do so because both sides committed before they understood the work. The Sprint is small enough to validate fit on both sides — and produces something real either way.
What You Walk Away With
- A written architectural assessment of your AI stack
- One shipped foundation piece — eval framework, architecture document, migration recommendation, or Claude Code rollout playbook
- Prioritized roadmap for the next phase
- Honest recommendation: extend into retainer, walk away with what we built, or referral if I'm not the right fit
Example Focus Areas
- Eval framework and observability for an AI feature you're afraid to change
- Architecture review of an agent loop that works in demos but breaks in production
- Claude Code team rollout: CLAUDE.md, MCP servers, shared skills
- AI stack consolidation and tooling migration recommendation
- Pilot architecture for a new AI feature you haven't built yet
Production AI Retainer
$15,000–$60,000/month
After the Sprint, you can extend into a retainer at the intensity you need. Engagements continue month-to-month at a flat rate with flexible hours, with direction reset every two weeks and the engagement evaluated at the quarter mark. No timesheets. No calendar-bound deliverables. Two-month minimum on retainer after Foundation Sprint, then 30-day notice either side. Specific monthly rate determined by scope and intensity — discussed on the intro call.
Advisory
5–10 hrs/week
Senior architectural guidance without heavy implementation. Architecture review, technical strategy, hiring support, async availability for high-stakes decisions.
Best for: You have a strong engineering team but need a senior voice on AI architecture, hiring, or strategic decisions.
Recommended
Embedded
10–20 hrs/week
Hands-on implementation alongside advisory. I'm in your codebase shipping foundation pieces while also helping you make architectural calls.
Best for: You want me writing code on the highest-leverage workstreams while still advising the broader stack.
Full
25–40 hrs/week
Senior IC running point on an AI workstream end-to-end. Build-heavy engagements where I function as your AI lead.
Best for: Build-heavy engagements where AI is the product or core differentiator. Foundation Sprint required first.
Every Retainer Includes
- Weekly working session and async Slack access
- Direction reset every two weeks; engagement evaluated at the quarter mark
- Written artifacts your team can reference long after I'm gone
- Flat rate, flexible hours within the stated range — no timesheets
- Two-month minimum after Foundation Sprint, then 30-day notice either side
What I Work On
Inside a retainer, we prioritize from a menu — not a fixed deliverables list. The work shifts as the codebase teaches us what's actually broken.
Production AI
- LLM integration: prompt design, structured outputs, tool use, streaming
- RAG pipelines that return relevant results, not just similar ones
- Agent architectures with working memory and tool orchestration
- Cost monitoring, evals, and observability for AI features
Claude Code & Agentic Workflows
- Claude Code team rollouts (CLAUDE.md systems, MCP servers, shared skills)
- Workshops on your actual codebase
- Adoption playbooks that move the team beyond a few power users
- Custom MCP servers and agent skills tuned to your stack
Technical Leadership
- Architecture review and technical strategy
- Hiring support for senior AI/engineering roles
- Engineering process and conventions
- AI stack hardening: consolidation, migration recommendations, cost controls
30-Day Money-Back Guarantee
If after Month 1 of any retainer (or end of the Foundation Sprint) you don't have a written architectural plan and at least one shipped foundation piece, full refund. No questions, no hard feelings.
I limit engagements to 1–2 active at a time specifically so I can deliver to this standard. The guarantee is honest pricing of my own confidence in the work.
Who This Is For
Good fit
- Teams shipping AI into production, not exploring possibilities
- Engineering orgs that need a senior technical voice without a full-time hire
- Founders who built a prototype and need help getting it to production
- Teams adopting Claude Code at scale and stuck on the rollout
- Companies in regulated or high-stakes environments where AI has to actually work
Not a fit
- Need extra engineering hands to grind tickets
- Looking for "add AI to everything" consulting
- Pure research or open-ended exploration without ship targets
- Want me to write your prompts. If it's a prompt problem, you don't need me.
What Clients Say
"He quickly understood where we were with AI tooling and gave us immediately actionable advice, not generic frameworks. He identified gaps we hadn't considered, walked us through how he architects agent loops in production, and helped us think through our product-level agent strategy without over-engineering it."
"I found his videos especially clear-headed. I booked a couple of private sessions to discuss OpenClaw, and to my delight, he was equally clear-headed in person."
Frequently Asked Questions
Do I have to start with a Foundation Sprint?
Strongly recommended. The Sprint is small enough to validate fit on both sides and produces something real either way. For Full retainer engagements ($50k/mo), the Sprint is required — too much is at stake on both sides to skip the validation.
How does the retainer actually work?
Flat monthly rate, flexible hours within the stated range. No timesheets, no calendar-bound deliverables. Direction is reset every two weeks; the engagement is evaluated at the quarter mark. Engagements continue month-to-month with a two-month minimum on retainer after the Foundation Sprint, then 30-day notice either side at any time.
What's the difference between Advisory, Embedded, and Full?
Hours and the mix of advice vs. implementation. Advisory is mostly strategic guidance with light hands-on work. Embedded means I’m shipping code on the highest-leverage workstreams while also advising. Full means I’m functioning as your senior AI lead, running point on the work end-to-end.
What's covered by the money-back guarantee?
If after Month 1 of any retainer (or end of the Foundation Sprint) you don’t have a written architectural plan and at least one shipped foundation piece, full refund. No prorating, no carve-outs, no fine print.
What if I just need a few hours of advice?
1:1 coaching sessions are $300/hour. See /coaching/. Coaching is for individual engineers and leaders working through a specific challenge — not the right fit for ongoing team-level work.
What's your availability?
I take 1–2 active engagements at a time. Response time within 48 hours. Foundation Sprints typically start within 2 weeks of signing.
Do you write code?
Yes — Embedded and Full retainers include hands-on implementation in your codebase. Advisory retainers are mostly strategic guidance with limited hands-on work. The Foundation Sprint always ships something concrete.
What's your background?
15+ years building production software. Previously CTO at Buoy Software where I scaled the engineering team from 0 to 50+ and shipped FDA-cleared medical device software. Currently Senior SWE at August Health (clinical software for senior living). I advise because I still build.
Do you work with non-Rails teams?
Yes. Rails is a strength, but production AI patterns work across stacks. I’ve shipped AI features in Rails, Node, and Python environments.
Let's Talk About Your Project
Book a free 30-minute intro call. We'll talk through what you're building and where you're stuck. I'll be honest about whether I'm the right fit and which engagement makes sense.
No pitch deck. No pressure. Just a conversation.