AI Research Engineer - Quant Finance, Low-Latency ML

P2P; New York (NY); On-site; USD 255,000 - 345,000; Full time

Job summary

P2P is seeking world-class engineers to collaborate with our research, trading, and engineering teams to build state-of-the-art ML systems for quantitative finance. You’ll optimize training pipelines, develop low-latency inference, and push AI research from concept to production.

Proficiency in Python and/or C++, experience with PyTorch/JAX/TensorFlow, and GPU programming (CUDA/Triton/SYCL/ROCm) are essential. CPT/OPT eligible; base salary is $300,000 per year.

Qualifications

• Proven ability to build and deploy ML systems at scale.
• Strong programming skills in Python and C++.
• Experience with PyTorch, JAX, or TensorFlow.
• GPU/accelerator programming experience with CUDA or similar.

Responsibilities

• Apply state-of-the-art ML techniques to complex domains.
• Collaborate with researchers and quants to build reusable AI/ML frameworks.
• Optimize training pipelines for HPC resources.
• Integrate AI/ML models into latency-sensitive production systems.
• Work across Python/C++/CUDA and other low-level GPU languages.
• Build observable, high-performance AI/ML systems.
• Other duties as assigned.

Skills

Python

C++

GPU programming

ML systems at scale

English communication

Collaborative work

Tools

PyTorch

JAX

TensorFlow

CUDA

Triton

SYCL

ROCm

Job description

P2P is seeking world-class engineers to collaborate with our research, trading, and engineering teams to build state-of-the-art ML systems for quantitative finance. You’ll optimize training pipelines, develop low-latency inference, and push AI research from concept to production.

Proficiency in Python and/or C++, experience with PyTorch/JAX/TensorFlow, and GPU programming (CUDA/Triton/SYCL/ROCm) are essential. CPT/OPT eligible; base salary is $300,000 per year.