XTX Markets

The program runs 12–14 weeks, is open to students in the middle years of an advanced degree (ideally PhD in computer science, electrical engineering, mathematics, or a related quantitative field, with at least one year remaining before graduation), pays a base of $35,000 per month, and requires four days a week onsite at the Hudson Yards office. Top performers are considered for full-time research roles in either the New York or London teams. Start dates are flexible across the year, and the firm is currently only accepting candidates available from fall 2026 onward. This is a rolling, conversion-oriented research program rather than a cycle-driven summer-analyst class, so a candidate watching XTX alongside a wider quant recruiting season is not racing a single autumn deadline but is competing in a deliberately small cohort.

Three things shape the early-career experience here:

• Research-scale compute. The firm runs an in-house cluster of over 25,000 GPUs with 650 petabytes of storage, and is investing over €1bn building a new data centre campus in Kajaani, Finland to expand that capacity further.
• Direct research-to-production pipeline. XTY Labs, the firm's machine-learning research arm established in 2024, exists to integrate models into live trading on a weeks-not-months cadence.
• PhD-weighted hiring. The published internship targets students in the middle of advanced degrees with a publication record at leading ML venues; the cohort is deliberately small and selective.

For a candidate choosing where to spend a research-heavy internship, the trade-off here is academic-style freedom and unusually large compute against a narrow domain (price forecasting for trading), and an unusually short loop from a published idea to live market deployment.