Campus AI Research Engineer – Deep Learning(full-time)

Jump Trading Jump Trading · Quant · Chicago, New York City · Front Office

Research Engineer at Jump Trading focused on applying state-of-the-art ML techniques to financial markets. The role involves implementing research projects from concept to production, optimizing training pipelines, integrating ML models into production systems with latency constraints, and building large-scale, observable, performant, and flexible ML systems. Requires strong publication record, ML background (deep learning, transformers, SSMs), and proficiency in Python/C++/CUDA. Experience with HPC, distributed training, and GPU optimization is a plus.

What you'd actually do

  1. Apply state-of-the-art techniques to complex and challenging domains.
  2. Work closely with researchers and quants to build flexible and reusable frameworks for financial ML.
  3. Optimize training pipelines to make the best use of our HPC resources.
  4. Integrate ML models into production systems where latency matters.
  5. Build large-scale ML systems that are observable, performant, and flexible. Help improve productivity by reducing the iteration cycle time on research.

Skills

Required

  • Strong publication record at ICML, ICLR, AAAI, NeurIPS, UAI, KDD, or equivalent and/or contributions to open-source AI research
  • Strong general ML background with exposure to modern deep learning techniques and/or language modeling architectures (e.g. transformers, SSMs)
  • Solid development skills in Python and/or C++
  • Familiarity with ML libraries/frameworks such as PyTorch, JAX, and/or TensorFlow
  • Intellectual curiosity, versatility, and originality combined with a pragmatic outlook
  • Ability to thrive in a collaborative, team-oriented environment
  • Ability to reason through quantitative problems and communicate effectively with trading researchers
  • Reliable and predictable availability

Nice to have

  • Experience with HPC and distributed large model training
  • Experience with GPU performance optimization (CUDA or ROCm)
  • Experience with end-to-end model development
  • Strong opinions on best practices in ML research, tooling, and/or infrastructure

What the JD emphasized

  • Strong publication record at ICML, ICLR, AAAI, NeurIPS, UAI, KDD, or equivalent and/or contributions to open-source AI research
  • implementing an open-ended research project from concept to production
  • continuously improving model design, tools, and infrastructure
  • latency matters

Other signals

  • implementing an open-ended research project from concept to production
  • continuously improving model design, tools, and infrastructure
  • apply machine learning to achieve state-of-the-art capabilities in complex and challenging domains