Research Engineer: AI Training Infrastructure & Innovation

Fireworks AI; New York (NY); On-site; USD 150,000 - 230,000; Full time

Job summary

Fireworks AI is seeking a Research Engineer to join our team at the intersection of model research and training infrastructure. You will design novel architectures, reproduce state-of-the-art results from literature, and build the distributed systems needed to scale research to production.

You will balance theoretical exploration with practical engineering across large GPU clusters. The role requires strong programming in Python/C++/Rust and expertise in ML frameworks (PyTorch/JAX/TensorFlow).

Qualifications

• Strong programming skills in Python, C++, or Rust.
• Deep practical knowledge of ML frameworks: PyTorch, JAX, or TensorFlow.
• Experience with large distributed systems and parallel computing (CUDA, NCCL, MPI).
• Strong foundation in linear algebra, calculus, probability, and statistics.
• Proven track record of implementing complex deep learning algorithms from scratch.

Responsibilities

• Conduct Open-Ended Research: explore new model architectures and optimization techniques; design experiments; iterate based on results.
• Reproduce and Extend State-of-the-Art: implement and reproduce results from recent ML papers; scale methods to larger datasets and models.
• Build and Scale Training Infrastructure: design, implement, and maintain high-performance distributed ML systems; optimize training loops and data loaders.
• Bridge Science and Engineering: translate mathematical concepts into robust, efficient code.
• Collaborate Cross-Functionally: work with Research Scientists to unblock experiments with tooling and hardware-aware co-design.

Skills

Python

C++

Rust

PyTorch

JAX

TensorFlow

Distributed systems

CUDA

NCCL

MPI

Linear algebra

Calculus

Probability & statistics

Deep learning algorithms

Education

Master's or PhD in Computer Science, ML, Physics, Mathematics

Tools

CUDA

Triton

Job description

Fireworks AI is seeking a Research Engineer to join our team at the intersection of model research and training infrastructure. You will design novel architectures, reproduce state-of-the-art results from literature, and build the distributed systems needed to scale research to production.

You will balance theoretical exploration with practical engineering across large GPU clusters. The role requires strong programming in Python/C++/Rust and expertise in ML frameworks (PyTorch/JAX/TensorFlow).