JPMorgan Chase CEO Jamie Dimon told Bloomberg on May 21 that the bank will hire more AI specialists than traditional bankers "in certain categories." Seventeen days earlier, Anthropic had announced a $1.5 billion joint venture with Goldman Sachs, Blackstone, and Hellman & Friedman designed to embed AI engineers inside hundreds of private equity-owned companies across the United States. New York, which has spent three years watching San Francisco absorb the majority of the frontier AI labor market, is now the primary target of two of 2026's most consequential AI workforce moves — and neither fits the standard remote-versus-office binary that has dominated the conversation since January.
What's Happening
JPMorgan Chase runs the largest AI engineering organization on Wall Street, and it is expanding. The bank's Machine Learning Center of Excellence employs more than 200 ML scientists, software engineers, and product managers in New York, with concentrations in large language models, NLP, time series, and reinforcement learning. JPMorgan's LLM Suite — an internal AI assistant modeled on ChatGPT — has been adopted by more than 200,000 employees, one of the largest enterprise LLM deployments in the country by volume. The bank committed a $19.8 billion technology budget for 2026, roughly 10 percent of revenue and up approximately 10 percent from 2025, with allocation heavily weighted toward AI agents, workflow re-engineering, and infrastructure. Lori Beer, who leads more than 65,000 technologists across the bank, directs the program.
Compensation for AI engineering roles at JPMorgan is not what the finance-is-below-tech narrative suggests. AI Engineers in New York earn total compensation of $194,000 to $328,000, with a median of $276,000, per Levels.fyi data. Machine Learning Engineers run $170,000 to $249,000 at the median of $195,000. Those numbers sit below the frontier lab bands — Anthropic's software engineers in San Francisco cleared a median of $665,000 in total comp per Levels.fyi data as of June 2026 — but they are structurally competitive with mid-senior roles at Microsoft's Redmond AI org, where AI engineers run a median of $274,000 to $282,000. For an AI engineer who wants to work in New York rather than San Francisco, JPMorgan is now a credible full-time destination rather than a fallback.
Goldman Sachs is building a parallel track. The bank is actively hiring Senior AI/ML Engineers in New York at base salaries of $150,000 to $300,000, with Applied AI Quantitative Engineer roles running $115,000 to $180,000 at base (per Goldman Sachs career portal and LinkedIn postings, Q2 2026). Goldman's Global Banking and Markets division is running active AI engineering campaigns that are production-grade in scope — not strategy-adjacent roles with AI in the title, but positions requiring LLM deployment, vector store management, and enterprise AI framework experience.
The more structurally significant move belongs to Anthropic. The $1.5 billion joint venture announced May 4 — backed at $300 million each by Blackstone and Hellman & Friedman, $150 million from Goldman Sachs, and additional capital from General Atlantic, Leonard Green, Apollo Global Management, Singapore's GIC, and Sequoia Capital — will embed applied AI engineers inside hundreds of mid-sized private equity-owned companies across the United States. The target verticals are healthcare, manufacturing, financial services, retail, and real estate. The operating model is specific: Anthropic engineers will sit alongside the venture's team to identify where Claude can have the most impact, build custom solutions, and support clients over the long term. This is not a software licensing arrangement. It is, structurally, a workforce placement program that distributes AI lab talent across the country at a scale no previous deployment model has attempted.
Why It Matters
The dominant narrative about AI talent geography for the past four years was simple: the people who do frontier AI work live within a few miles of San Francisco, and if you wanted to build in that space, you relocated or you missed out. The frontier labs' office-first policies reinforced this. Anthropic carries fewer than 8 percent remote-eligible roles on its job board as of June 2026; OpenAI sits at 4 percent (per ENTRA Job Signal Index, Q2 2026). The concentration play produced density that the labs argue is irreplaceable for safety research and model development. It also produced a single-city bottleneck that the broader market is now routing around.
Dimon's Bloomberg statement is not a rhetorical positioning. JPMorgan is already operating one of the largest enterprise LLM programs in the country, with 200,000 active internal users and a $19.8 billion technology budget behind it. When the largest US bank explicitly reorients its stated hiring priorities toward AI specialists over bankers, it announces a durable institutional competition for the senior ML engineering pool — a competition anchored in New York, not San Francisco.
Anthropic's joint venture does something more structural: it creates a class of AI engineers whose primary work context is geographically distributed across hundreds of client sites. These are Anthropic-trained engineers, operating inside the company's deployment model, working from PE-owned companies across the country rather than from Howard Street in San Francisco. That is a form of distributed AI labor that no standard taxonomy of "remote" or "hybrid" captures. The venture's built-in client pipeline — Goldman, Blackstone, Apollo, General Atlantic and their respective portfolio companies — means Anthropic effectively has a structured rationale to hire senior applied engineers who are geographically proximate to those institutions. Many of those institutions are in New York.
There is a compensation spread worth watching. Senior AI researchers at Anthropic earn $746,000 to $1.05 million in total comp per Levels.fyi data from June 2026. AI engineers at JPMorgan earn $276,000 at the median. The delta is real. But the question for an AI engineer in 2026 is not only about the absolute number. It is about location, career optionality, and whether the work is primarily research or deployment. A candidate who will not or cannot relocate to San Francisco now has multiple paths into applied AI work in New York that did not exist two years ago — including a path that carries an Anthropic employment relationship.
What's Next
Three dynamics will define the Wall Street AI talent market through the end of 2026.
The embedded engineer model will face its first staffing test. Anthropic's joint venture requires AI engineers who can manage client relationships, navigate institutional culture, and build production-grade AI systems simultaneously. That profile does not exist in large numbers at any lab. How the venture recruits, trains, and retains the engineers it places at client sites will determine whether the model scales or collapses under the weight of deployment complexity. Hiring engineers willing to work embedded at financial and PE-owned firms — rather than on a research campus — is the core organizational challenge, and the geographic bet embedded in it.
JPMorgan's AI headcount will draw direct competition from Meta and OpenAI's New York presences. Meta Superintelligence Labs operates a New York office alongside its Menlo Park headquarters — part of a distributed footprint that spans Menlo Park, London, Paris, Seattle, Pittsburgh, Tel Aviv, and Montreal. OpenAI maintains a Manhattan team that services enterprise clients across financial services, legal, and healthcare. Senior ML engineers in New York in H2 2026 will see simultaneous offers from JPMorgan ($276,000 median total comp), Goldman Sachs ($150,000–$300,000 base), Meta MSL (research engineer base of $154,000–$217,000 plus equity on a significant valuation, per ENTRA Talent Index Q2 2026), and the Anthropic embedded venture program. That four-way competition for a narrow candidate pool is new in New York's AI market.
Goldman Sachs's co-investor position in the Anthropic venture is strategically underweighted in public analysis. The bank committed $150 million — the smallest check among the anchor investors — but its role as a potential client, institutional partner, and talent pipeline for the venture's engineering cohort is more significant than the capital figure suggests. If the Goldman partnership deepens into a staffing relationship — Goldman AI engineers rotating into the venture, Anthropic engineers gaining direct exposure to trading and asset management workflows — New York will have its first institutional AI talent circulation loop outside the Bay Area.
For three years, New York has been the city where AI companies opened satellite offices to be near enterprise clients. The Dimon declaration and the Anthropic venture mark something different: the city where the AI hiring decisions are being made, not just serviced.
For context on how the broader US AI hiring market shifted across H1 2026, see the ENTRA US AI Hiring Midyear Analysis. For a deep dive on Anthropic's talent strategy behind its post-Series E expansion, see Anthropic's Talent Machine.
Global Career PlatformFind AI talent. Find your next role.
Booking is hotels. · Airbnb is apartments. · ENTRA is global careers.