Work

Tier 1

Enterprise AI Implementation

You might need this if you have a board-approved AI mandate and no one inside the building has shipped a production model end-to-end.

• Reduced model-evaluation cycle from six weeks to nine days while keeping a written audit trail.
• Shipped the first production agent deployment with a documented rollback path and a named on-call rotation.
• Closed the gap between board-deck AI promises and operational AI by retiring two stalled vendor pilots.

A two-quarter implementation engagement. James embeds with the platform and ML teams as the senior practitioner of record — architecting evaluation harnesses, agentic workflows (multi-step LLM systems with tool use), and retrieval-augmented generation (RAG; grounding model output in your own data) where the buyer-visible outcome warrants it. We use the buyer's existing cloud and identity stack; we do not introduce new vendors. Governance is a deliverable, not a slide.

Starts in the low six figures per quarter; scoped against named outcomes, not headcount.

Two on-site visits within 90 minutes of San Diego per month, included.

Tier 2

AI Strategy Sprint

You might need this if your leadership needs an operational AI plan, not another consulting deck.

• Replaced a 60-page strategy deck with a one-page memo and a six-month sequenced roadmap of named workloads.
• Identified two production-ready AI workloads with measurable revenue impact and three to defer indefinitely.
• Aligned legal, security, and product on a written model-risk policy, signed by the executive sponsor.

A four-to-six-week senior-led strategy sprint. Outputs are a written one-page strategic memo, a sequenced workload roadmap, and a model-risk policy draft. James interviews engineering, security, legal, and product leadership directly; recommendations are vendor-neutral and grounded in what the company can operationally support. The Model Context Protocol (MCP; an emerging standard for tool-use orchestration across LLMs) is evaluated where it shortens time-to-value.

Mid five figures, fixed-fee. One memo, no slides.

Tier 3

Technical Due Diligence

You might need this if you are acquiring or investing and need a senior practitioner to read the model code, not just the pitch.

• Surfaced two undisclosed model-quality liabilities before close and adjusted the deal terms accordingly.
• Confirmed the target's evaluation harness was real and reproducible, materially raising buyer confidence.
• Produced a 90-day post-close integration plan that the acquirer's CTO accepted without revision.

A two-to-four-week technical diligence engagement scoped to a single transaction. James reads the actual repository, runs the evaluation harness, interviews the target's senior practitioners, and writes a confidential one-page memo to the deal team — risks, mitigations, and a realistic 90-day integration plan. Vendor-neutral; we do not sell the target a follow-on.

Engagement scoped per transaction; mid five to low six figures depending on target complexity.

Tier 4

Fractional CAIO

You might need this if you need senior AI judgment in the room two days a week, not a full-time hire you can't justify.

• Held the senior-AI-leader seat in board, customer, and investor meetings without taking the full-time line item.
• Hired the company's first two ML engineers and stood up the on-call ML reliability rotation in the first quarter.
• Replaced a stalled vendor relationship with an in-house model deployment running on the existing platform stack.

An ongoing fractional Chief AI Officer engagement, typically two days per week, billed against a six-figure annual envelope. James is the named senior AI leader of record — internal, external, regulatory — until the company is ready to hire a full-time successor. Includes hiring rubric, governance authoring, and a clear handoff plan from day one.

Fractional retainer; six-figure annual envelope. Capped at two concurrent fractional engagements per year.

Tier 5

Workshops & Exec Ed

You might need this if your executives and engineers need a working vocabulary, not a vendor demo.

• Aligned a 40-person executive team on a shared, accurate AI vocabulary in one half-day session.
• Equipped the senior engineering bench with a working evaluation rubric for incoming AI vendors.
• Delivered a written reading list and decision framework that survived past the workshop date.

Half-day, one-day, or two-day senior-led workshops for executive teams or senior engineering benches. On-site or remote. Materials are written documents, not slide decks. Each workshop ends with a decision rubric the team owns and uses inside the next budget cycle.

Five-figure fixed fee per workshop; on-site within Southern California included.