Anthropic launches Claude Frontier Academy, a $100 million training program designed to produce 10,000 Frontier Deployed Engineers by the end of 2027, focusing on enterprise AI engineers.

Cohorts are currently in San Francisco, New York, and London, with plans to scale to 10,000 engineers by year-end 2027.

The academy targets engineers from Anthropic's Claude Partner Network, aiming for first certifications in early 2027 and ongoing credentialing through the network.

Participation is by nomination, and organizations can verify eligibility through their Anthropic or Partner accounts.

Industry observers see this as a shift from model development to deployment ecosystems, signaling a battleground in enterprise AI amid regulatory and competitive uncertainties.

Early quotes from firms like Accenture, Capgemini, McKinsey, Commonwealth Bank of Australia, and Novo Nordisk highlight anticipated benefits of deploying Claude and expanding AI capabilities.

The move positions Anthropic against OpenAI’s enterprise push by offering embedded, credentialed personnel rather than licensing alone.

Anthropic reports strong financials and large compute commitments, while facing regulatory scrutiny and industry pushback tied to antitrust concerns and safety standards alignment.

Selected fellows will spend a year working with nonprofits across the U.S., including YMCA and Goodwill.

Frontier Engineers act as customer-facing technologists who tailor AI to a company’s data and processes, influence what to build, and drive adoption.

Fellows receive an $85,000 annual stipend plus benefits, mentorship, professional development, and a Claude token budget.

The program includes multi-day in-person training, simulated deployments, and a 12-week residency, with the first credentials—Claude Resident Engineer and Claude Frontier Deployed Engineer—expected early 2027.