Quant firms court AI PhDs with $800,000 offers and academic funding

Quantitative trading firms are offering machine-learning PhDs first-year packages that typically run $450,000 to $650,000 and can reach $800,000, according to one recruiting firm. The AI labs and startups chasing the same researchers are selling something else: prestige, visibility and, in some cases, a stake in a fast-growing technology company.

The response from some of finance's most secretive employers has been to spend in public, on people who may never work for them. G-Research, a London-based firm of about 1,000 people, announced this summer the largest private-sector investment in the European Laboratory for Learning and Intelligent Systems, a network of machine-learning researchers, with funding for 36 of them, Bloomberg's Justina Lee reported Oct. 27. It also funds 43 PhD students across eight British universities. Charles Martinez, who leads the firm's academic outreach, said there is no expectation that the sponsored researchers will join G-Research or do work applicable to finance. The payoff, he said, is reputational: a higher profile among potential recruits.

G-Research is not alone. Qube Research & Technologies has launched QRT Labs, which sponsors more than 70 PhD and postdoctoral researchers. Jump Trading funds academic fellowships, and Abu Dhabi Investment Authority has started a quant research residency for PhDs. Capital Fund Management, the Paris quant fund with roughly $35 billion, hired a New York University math professor to run a machine-learning lab with two full-time researchers and a rotating group of postdocs. "They are not under the pressure to build alpha models," said Laurent Laloux, CFM's chief product officer.

The structural disadvantage is one the firms acknowledge. Quant research mostly stays private, and Martinez described the conversation his recruits can expect with peers at an AI lab: "Your friend in OpenAI will be, 'OK, I did this, I did that,' and you'll be like, 'Yeah, I did something, but I can't tell you.'" His counter-pitch is that G-Research's quants run their groups the way a university runs a PhD group, with one of the most powerful computing resources in the world behind them.

Compute has become a recruiting argument in its own right. Hudson River Trading, XTX Markets and Jane Street operate their own GPU-packed data centers. "It isn't just about compensation," said Tom Chambers, a partner specializing in AI at recruitment firm Maven Partnership. "It's about commitment to compute and infrastructure." At NeurIPS, a leading machine-learning conference, nearly half of the top-tier sponsors for this December's event are financial firms.

People are moving in both directions. HRT, known for its longstanding AI lab, has lost several employees to OpenAI, and Two Sigma quants have gone to Anthropic. A few Google DeepMind scientists have joined Jump and Citadel Securities.

At Jump, Chief Technology Officer Alex Davies said the firm has always had academic partnerships but decided to publicize them because younger researchers are more likely to investigate potential employers themselves than to go through recruiters. "AI labs are clearly the biggest competitor right now," said Davies, a 15-year Jump veteran. Some Jump employees now carry the title "member of technical staff," the deliberately vague label popularized by OpenAI and Anthropic, which Davies likes because it erases the line between researchers and engineers.

The hiring documents reflect the same pitch. A posting on IMC Trading's careers page for a London-based principal research scientist in machine learning targets sitting academics, from assistant professors to principal investigators, states that prior finance experience is not required, and promises representation at NeurIPS, ICML and ICLR. Publication is offered "where appropriate and consistent with intellectual property considerations." A separate listing posted Oct. 22 through eFinancialCareers, for an unnamed London systematic investment firm, invites candidates finishing a PhD within the next year and says outright: "Financial experience is not required."