How I can help
Five engagement shapes. Each one starts with evidence, ends with something your team can run without me, and is scoped before any work begins.
- S-01
Production AI Systems / Forward Deployment
For companies with an AI initiative that needs to become a production system. I work from architecture through implementation and launch: model integration, APIs, data, infrastructure, evaluation and deployment.
Fixed-scope build or monthly retainer, scoped together first
You get
- An architecture your team can own, with the decisions written down
- The hard parts built and shipped: AI features, integrations, data pipelines
- Cloud infrastructure, CI/CD and a staging gate before production
- Evaluations and rollback, so the next change is safe too
I need from you
- A product owner who can make decisions
- Access to your repositories, cloud accounts and data
- A launch target, even a rough one
- S-02
Architecture Drift Audit
For teams whose architecture decision records, design docs or runbooks no longer match the code, often after a year of fast shipping or heavy use of AI coding agents. Call me before a migration, an acquisition review or a compliance audit.
2–3 weeks, fixed scope
You get
- A verdict on every decision record: drifted, current or status wrong
- Each verdict traced to quoted evidence from your code, schema and infrastructure
- Run receipts: pinned commit, models, budgets and evidence hashes
- A ranked list of the drift that matters, with suggested fixes
- A walkthrough with your engineers
I need from you
- Read access to the repository and decision records
- Read-only access to schema and infrastructure config, or exports of them
- One engineer who can answer questions about intent
- S-03
Agent Reliability / Harness Build
For teams with agents that work in the demo and wobble in production. I build the harness around them so every run is evidenced, checked and reversible, and a wrong answer is caught before it reaches a user.
4–8 weeks per agent workflow, fixed scope
You get
- Deterministic evidence gathering, frozen and hashed before the model sees it
- Schema-bound model calls with a validator that decides what counts
- Per-run receipts, token and cost budgets, and one-command rollback
- A release gate: a test suite the agent must pass before it ships
- Handover docs and a session with your team
I need from you
- Access to the agent's code and a non-production environment
- Real examples of good and bad runs
- A technical owner on your side who will keep the harness after handover
- S-04
AI Evaluation Design
For teams that can't say whether the last prompt, model or fine-tune made things better. I design evals that answer that question honestly, including ones that are hard to game.
2–4 weeks, fixed scope; optional re-runs on retainer
You get
- A pre-registered eval plan: metrics, sample and pass bar fixed before scoring
- A sealed, hand-labelled answer key
- Regression and replay suites wired into CI
- Reward and eval design for RL or fine-tuning runs, where relevant
- A written read-out with honest limits
I need from you
- The task definition and real (or realistic) inputs
- Someone with domain knowledge to label or review the answer key
- API or compute budget for the eval runs
- S-05
Fractional AI Lead
For founders and engineering leaders who need senior AI judgment and hands-on delivery without a full-time hire. I set the architecture and the agent-team process, then write code alongside your engineers.
Monthly retainer, a fixed number of days per week
You get
- Architecture decisions, written down as decision records
- An agent-assisted engineering process with review gates
- Hands-on delivery of the hardest parts
- Customer-facing technical work: discovery, reviews, go-live
I need from you
- A clear owner I report to
- Access to your repositories, cloud accounts and team channels
- A standing weekly review
Terms of engagement
- Contract or retainer, invoiced through Syndicate Digital Productions Inc.
- Fully remote, from Vancouver, BC
- NDA-friendly: client work is described by shape and outcome, never by name
- Works directly with your team or as a subcontractor to your agency
Need a senior AI engineer who can own delivery?
Tell me what you're building and where it breaks. I reply to every message myself.
Contract CTO · Principal AI engineering · Forward-deployed AI engineering · Enterprise AI architecture