Anthropic announced on October 2, 2026 that it is putting $100 million into training 10,000 Frontier Deployed Engineers (FDEs) by the end of 2027. The pitch behind Claude Frontier Academy is simple: the bottleneck in enterprise AI isn’t model capability, it’s the small pool of engineers who can take a model from idea to a system actually running inside a business.
Why a training program, not another certification
The pattern Anthropic describes will be familiar to anyone who has shipped AI inside a large company: a handful of deeply skilled people produce most of the value, and everyone else waits on them. Steve Corfield, Anthropic’s Global Head of Business Development and Partnerships, frames it as high-agency engineers with Claude access plus real understanding of the business being able to transform how the company runs.
So instead of a course catalog, the entry point is the Frontier Deployed Engineer Residency, structured on the medical model — learn from practitioners, train on realistic cases, get assessed before practicing alone. That’s a deliberate contrast with volume-based certification. If you’ve followed Anthropic’s Claude rollout at scale, like what we covered in the Barclays enterprise rollout, you know the hard part was never API access; it was people who could carry a use case through security review and into production.
How the residency actually works
The mechanics matter for builders evaluating whether this applies to them:
- Organizations nominate engineers who already have strong fundamentals and LLM building experience, each arriving with a named Claude project to lead on return. Prior agent-building experience is not required.
- Phase one is a multi-day in-person program with Anthropic engineers and licensed instructors: a simulated enterprise deployment from use case selection through security review to handover, ending in a graded practical. Passing earns the Claude Resident Engineer badge.
- Phase two is a 12-week residency leading a real Claude use case at the engineer’s own organization, with Anthropic support. A second assessment leads to the Claude Frontier Deployed Engineer badge, with the first expected in early 2027.
The detail I find most telling is that everyone arrives with a named project. This isn’t training divorced from delivery — the 12-week residency is the delivery, inside the learner’s own company.
Who is in the first cohorts
Cohorts are already running in San Francisco, New York and London, drawn from Accenture, Bain, Capgemini, Commonwealth Bank of Australia, Deloitte, McKinsey, Morgan Stanley and Novo Nordisk. The mix is worth noting: consulting firms want Claude-ready bench talent for client work, while Commonwealth Bank says its teams produced up to 3x more code changes in the past year and sees the residency as a way to go deeper. Novo Nordisk ties it to lab-in-the-loop drug discovery workflows.
This builds on the Claude Partner Network, where professionals at 46,000 firms have earned more than 175,000 Claude certifications and nearly 4,000 people have completed Basecamp, Anthropic’s immersive onboarding program. The Academy targets the people who lead, not just the people who use.
What it signals for your own team
You don’t need a nomination slot to take something from this. The residency’s shape — one high-stakes named project, real security review, assessed against realistic scenarios rather than quizzes — is a reasonable template for any internal AI enablement program. Pick your most important workflow, assign an engineer with real fundamentals, and assess on the shipped system, not course completion.
One caveat: participation is nomination-based through Anthropic account or partner teams, so access is effectively gated to enterprises already in the ecosystem. For everyone else, the practical move is the same discipline without the badge — define the project, run the deployment gauntlet, and judge the result in production.