Hardware Machine Learning Engineer

About this role

Join a team deploying machine learning directly onto custom hardware in Chicago or New York. You'll architect ML solutions from the ground up, collaborating across traders, researchers, and engineers while having full control over the entire stack—from hardware to software—to solve some of the most computationally demanding problems.

What you'll do

• Co-design ML models with traders and quant researchers, incorporating hardware constraints as core design inputs
• Translate ML model requirements into custom hardware architectural decisions and roadmap priorities
• Implement and deploy ML inference solutions from prototype through production with hardware engineers
• Evaluate emerging research in neural architecture search, quantization, and ML systems for practical impact
• Optimize models for latency, resource utilization, and numerical precision on FPGAs or ASICs

What they're looking for

• Hardware design trade-offs (pipelining, resource utilization, fixed-point arithmetic)
• VHDL/SystemVerilog, HLS tools, or ML-to-hardware frameworks (hls4ml, FINN, Vitis AI)
• Neural network architectures and inference optimization
• Quantization techniques and ML frameworks (PyTorch/TensorFlow)
• Python and C++ for tooling and simulation
• Cross-disciplinary collaboration and communication
• ML compiler infrastructure (MLIR, TVM, XLA preferred)
• Latency-sensitive systems or high-frequency trading experience (preferred)