Latest Interviews
Showing 1–5 of 5 transcripts.
Clear all filters- Jane Street16 min
Dwarkesh Goes Inside Jane Street's Latest AI Data Center
Dwarkesh, Ron Minsky, Daniel Pontecorvo, Mark Mirchandani
Jane Street transformed a Texas data center into a high-density liquid-cooled facility housing 4,032 GPUs to execute large language model training and custom trading architectures. The retrofit replaces legacy air-cooling with an 18°C fluid distribution system that manages 140 kW per cabinet while utilizing proprietary software to dynamically redistribute power and prevent breaker trips during peak loads. Engineering safeguards now focus on mitigating new liquid-cooling risks like biological growth and leaks, enabling sub-100-nanosecond latency required for modern algorithmic trading compared to the millisecond scales of its historical "Hive" cluster.
- Jane Street59 min
The Thermodynamics of Trading with Daniel Pontecorvo
Daniel Pontecorvo, Ron Minsky, Dan Pontecorvo, Mark Mirchandani, David Abelson
Jane Street's physical engineering team manages a complex lifecycle of data centers and offices, prioritizing microsecond-level latency constraints and evolving from traditional air cooling to high-density liquid immersion solutions for AI infrastructure. The organization deploys proprietary monitoring stacks to prevent catastrophic failures while optimizing Power Utilization Efficiency through economizer cycles and rear-door heat exchangers that handle rack densities up to 170 kW. Their human-centric workspace design leverages modular furniture and circadian lighting to sustain collaboration, supported by a hiring strategy that blends industry veterans with graduates to challenge established engineering assumptions.
- Jane Street1h 20m
Building Tools for Traders with Ian Henry
Ian Henry, Ron Minsky, Mark Mandelbaum, Mark Mirchandani, David Eastman, Brian Dorsey
Ian Henry, a Jane Street engineer who transitioned from general web development to options trading tools, leverages the firm's custom OCaml ecosystem to build high-density interfaces that eliminate mouse usage and manage complex volatility surfaces. His recent work includes creating Bauble, a live 3D graphics environment in the Janet language that demonstrates how side projects using image-based state serialization and novel macro systems can rapidly inform internal production tooling. Through this process, Henry illustrates how Jane Street bridges the gap between low-latency exchange architecture and expert user needs by prioritizing type safety, extreme information density, and a parallel software universe free from standard web protocols.
- Jane Street1h 6m
The Uncertain Art of Accelerating ML Models with Sylvain Gugger
Sylvain Gugger, Ron Minsky, Jeremy Howard, Mark Mandelmann, Mark Mirchandani, Francesc Campoy, Gabriel Sanchez
Former fast.ai co-author Jeremy Howard discusses his transition from mathematics education to optimizing machine learning infrastructure at Jane Street, highlighting breakthroughs in learning rate schedules and image resizing that previously secured top benchmark placements. He details the development of the Hugging Face Accelerate library, a lightweight tool designed to abstract complex hardware parallelism and eliminate boilerplate code for training across diverse GPUs and TPUs. The discussion further explores the architectural constraints of financial data, the dominance of PyTorch's iterative execution model, and Jane Street's rigorous approach to reproducibility and custom model development for high-frequency trading.
- Jane Street37 min
A Jane Street Software Engineering Mock Interview with Grace and Nolen
Grace, Nolen, Nolan, Emily Fortuna, Colton Ogden, Todd Kerpelman, Jen Person, Colt McAnlisley, Dan Galpin, Mark Mirchandani
Nolan and Grace, experienced Jane Street employees, co-created a mock interview simulation featuring a unit conversion challenge that evaluates a candidate's graph-based algorithm design and problem-solving collaboration. During the session, the candidate developed a breadth-first search solution to minimize floating-point errors while iteratively correcting structural flaws regarding bidirectional graph construction and edge cases. The exercise underscores Jane Street's emphasis on code clarity, effective communication, and the ability to refine logic through dialogue rather than demanding immediate optimization or perfect syntax.