Anthropic invests $100 million to train 10,000 engineers by the end of 2027
The company is launching the Claude Frontier Academy with first cohorts already underway in San Francisco, New York and London — a nomination-based program designed to fill one of enterprise AI's most in-demand roles, weeks before an expected IPO.
Anthropic announced on Friday 2 October 2026 that it is investing $100 million in the Claude Frontier Academy, a program that will train 10,000 so-called "frontier deployed engineers" drawn from companies in the Claude Partner Network by the end of 2027. The figures come from Anthropic's own press release, as reproduced by CNBC, and have not been independently verified.
The first cohorts are already underway in San Francisco, New York and London, with engineers from Accenture, Morgan Stanley, Novo Nordisk and a range of leading consultancies, according to the press release (CNBC).
Who gets in, and how does the program work?
Entry is by nomination only. Anthropic says it is looking for hands-on software engineers with strong fundamentals and documented work with large language models. Prior experience with AI agents is not a requirement — but every participant comes with a named project they will lead when they return to their employer (The Daily Star).
The training itself is modeled on medical residency, according to Anthropic's own blog description as reported by IT Pro. Participants first complete a simulated enterprise deployment with a graded practical. Those who pass receive the "Claude Resident Engineer" badge and move on to a 12-week stint leading a real Claude implementation in their own organization, supported by Anthropic's own engineers (IT Pro).
The sources are not fully consistent on the certification details: CNBC writes that certification is expected "in early 2027" without naming the sequence, while The Daily Star describes a two-part badge ladder. The complete badge structure is therefore unconfirmed.
Why forward deployed engineers?
The backdrop is a pronounced imbalance in the labor market. The analytics firm Draup found that job postings for forward deployed engineers (FDEs) rose to 1,404 across five major consulting firms in the year ending in May — even as total job postings at seven firms fell roughly 24 percent year over year (Business Insider). Draup data provided to CNBC also shows that postings for FDEs and other AI-related roles have surged in the finance sector this year.
The role is, in short, about engineers who work closely with the customer's organization to get AI systems actually into production — work that models alone do not perform. In December 2025, Accenture had already established its own Anthropic Business Group and pledged to train around 30,000 of its own staff on Claude, including the firm's "reinvention deployed engineers" (Business Insider).
An ecosystem play, not classic corporate training
The academy should be read alongside Anthropic's broader push on certification and implementation services. According to Anthropic's figures, as reported by CRN, the company has now issued more than 175,000 Claude certifications to professionals at 46,000 companies — up from around 85,000 certifications at nearly 2,000 partner companies in August. Around 4,000 people have completed the Basecamp program (CRN).
Steve Corfield, Anthropic's global head of business development and partnerships, framed the effort as a market-building project in an interview with CRN: "The skills gap is not a partner problem or a customer problem. It's a market problem. So we want to build the market that can be sustained by this work we're doing."
The launch also comes just months after the start of Ode, a $1.5 billion joint venture for enterprise AI services, with backers including Blackstone and Goldman Sachs (IT Pro).
The demand has a documented foundation: OECD analysis shows that 40 percent of employers in manufacturing and finance cite skills as the main barrier to AI adoption, and that over half of SMEs do not use generative AI (AI Magazine).
The IPO context — with caveats
The launch comes at a time when Anthropic's finances have drawn heavy attention. Reuters reported earlier this week, citing a leaked IPO prospectus, that the company took in nearly $4.6 billion in revenue last year with an operating loss of more than $8 billion. It has also been reported that the company may be targeting a $2 trillion market valuation at listing. These figures stem from a document that is not publicly available and are unconfirmed — they should be read as reported context, not documented fact (CNBC).
Open questions
Several things remain unresolved. It is not known how the $100 million will be allocated, or how progress toward the 10,000-engineer target will be measured or verified. The details of the certification ladder and the complete list of partners in the first cohorts vary between sources. And all the program figures — the investment, the target, the ecosystem numbers — come from Anthropic's own statements, not independent verification. Whether the academy actually closes what Corfield calls a market problem is something only time and third-party data can show.