Engineering team track · In your company, on your stack

Build AI systems that survive production.

For the whole team: PMs, tech leads, juniors, seniors, frontend. Titles matter less than they used to. The mindset is the training.

Vero AIReal-time voice agents, built at CME Offshore and Siren Analytics alongside a great team
Strategy& / PwCAI systems for enterprise clients. Specifics under NDA
Viral AlchemyMy own build: a system that maps what is about to trend

What the training covers

What your team leaves with.

01

Agent harnesses

How to structure agents that do real work: tools, guardrails, recovery, orchestration. Vendor-neutral by design, so the skills outlive any single model or platform.

02

Context engineering

What the model sees decides what you get. Retrieval, memory, and prompt structure, all engineered.

03

Tokenomics

The economics of every call. Caching, routing, batching, model selection. Your team knows what a feature costs before it ships.

04

Production reliability

Failure modes, testing, and monitoring for systems built on models. The demo is where the real work starts.

How it runs

On your systems, not on slides.

A stalled pilot keeps billing every month it sits there. The budget is gone either way. This training runs inside your company, on your actual stack and your actual backlog. Sessions build working systems with your team in the room. The cost math stays on the table the whole time.

I learned the method on Vero AI. The voice chain half-worked, then broke, for weeks. The fix was never a hero move. It was pen and paper, the system drawn and redrawn, every piece tested alone, then connected one layer at a time. That is how the agent finally picked up the phone. That method is what your team gets.

Three sessions, three hours each. Cohort sized to your team. Pricing follows cohort size: ten people is not priced like fifty.

How it starts

One call. Then you decide.

Most teams are not choosing between me and another vendor. They are choosing between doing this and doing nothing. Doing nothing has a price too. The other options are a hire you cannot fill for months, or a consultancy that leaves a deck. Neither leaves capability behind.

Step 1

A 20-minute call

You describe where AI is stuck in your team. If I am not the right person, I say so and point you toward what would actually help.

Step 2

The Tokenomics Audit

A paid, fixed-scope review of your AI spend and pipeline, line by line. It ships as an interactive report: costs mapped, fixes ranked. You keep it whether or not we go further.

Priced on the call, scoped in writing before anything starts.

Step 3

Build and train together

We fix the system with your team in the room, so the capability stays when I leave. Your team's time is the real cost. Both numbers go on the table before we start. The goal is that you stop needing me.

The first step

Twenty minutes. You describe where AI is stuck.

You leave the call knowing whether I can help, roughly what it would take, and what it would cost to find out for sure. If I am not the right person, I will say so on the call.

The 20 minutes, in order

  1. You talk first. Where AI is stuck, what has been tried, what it costs today.
  2. I answer plainly. Whether I can help, and what I would look at first.
  3. You leave with a next step. An audit scope, a pointer elsewhere, or a clean no.

Before you book

  1. Your stack is not too messy to start. Messy is the normal starting condition.
  2. Training that does not survive the week is the normal outcome. These sessions build on your backlog and ship something real.
  3. You do not need budget approved to take the call. You need it approved to start step two.

The build log

What I am building, what it costs to run, and what broke on the way. Monthly, at minimum. Leaving takes one click.

[Email provider pending: Kit]