Edison AI
The field notes

Getting AI out of the demo and into production.

Short, practical essays on the role that barely existed two years ago, the method that gets AI live, and the stack that runs it. Written for operators, not spectators.

The role · 7 min read

The forward-deployed engineer: the highest-leverage AI job nobody trained you for

Demand for the role grew fast and the talent pool is nearly empty. What an FDE actually does, why companies are paying a premium, and why the usual paths don't produce one.

Read the essay →
The thesis · 6 min read

The last mile is where AI deployments die

The demo worked and the project still died. The four things that actually kill AI projects, and why none of them are about the model.

Read the essay →
The method · 7 min read

How to scope an AI deployment so it actually ships

Most projects are scoped to fail before anyone writes code. How to discover the real problem and cut the first version down to something you can ship in weeks.

Read the essay →
The stack · 8 min read

Connecting Claude, GitHub, and AWS: the operator's stack

Three tools, each doing one job well, connected in a loop you understand. The practical stack for turning an AI idea into a running system.

Read the essay →
Validate · 6 min read

How to validate an AI deployment before you trust it

"It seems to work" is not evidence. How to test on real inputs, set the bar in advance, and build the safety net that earns trust.

Read the essay →
Operate · 6 min read

What to do the day after go-live

Launch is not the finish line. It's the moment the system starts telling you the truth. How to operate a deployment so it survives real users.

Read the essay →
Start here

Get the free Forward-Deployed Engineer Starter Kit

The prompts, the stack, and the first steps of the method, sent straight to your inbox. The fastest way to see if this path is for you.