Anthropic Commits $100M to Train 10,000 AI Engineers

On 2 October Anthropic announced Claude Frontier Academy, a training programme backed by $100 million. The goal is to train 10,000 engineers by the end of 2027 to take AI into production inside a company, which is exactly where many projects stall.

What a “Frontier Deployed Engineer” is

That is Anthropic’s name for the profile it wants to train: an engineer who does not just use a model but can build a system of agents that solves a specific business problem and runs reliably. It is the same standard the company holds its own customer-facing engineers to.

How the residency works

The main programme is the Frontier Deployed Engineer Residency, in five stages:

1. Several days of in-person training with Anthropic engineers and instructors.
2. An exercise simulating an enterprise AI deployment.
3. A graded practical assessment.
4. A 12-week residency leading real Claude projects inside the participant’s own organisation.
5. A final assessment that awards the credential.

The first cohorts run in San Francisco, New York and London, with engineers from Accenture, Bain, Capgemini, Commonwealth Bank of Australia, Deloitte, McKinsey, Morgan Stanley and Novo Nordisk. The first credentials are expected in early 2027.

Who can join

There is no individual sign-up: organisations nominate their engineers. Anthropic is looking for strong software engineering fundamentals, experience building with language models and a habit of helping others adopt AI. Prior experience with agents is not required.

Why it matters

An AI company spending $100 million on training people rather than models says a lot about today’s bottleneck: models already do a great deal, but there are too few professionals who can integrate them well. If you work in tech, that profile will be in high demand whether or not you go through this academy. To start on your own, see our agent design patterns and the list of free courses.