Machine Learning Engineer (Hardware)

We are a global trading firm built on a strong research culture and advanced technology platform. Our teams across the US, Europe, Asia Pacific, and India collaborate to provide liquidity in financial markets. We focus on innovation, rigorous thinking, and continuous improvement across both our trading strategies and the systems that support them.

We are deploying machine learning directly onto custom hardware — and we’re looking for someone to help build it from the ground up. This is a rare opportunity to architect systems from scratch, influence research direction, and see your work drive real impact in one of the most demanding computing environments.

We own the full stack, hardware, software, and infrastructure, so when you encounter a bottleneck, you can solve it directly. There are no external dependencies or layers you can’t access. If you’re motivated by pushing the limits of performance and efficiency, this role offers that scope. Candidates from a range of backgrounds are welcome; prior trading experience is not required.

Your Core Responsibilities

• Co-design ML models with traders, quant researchers, and engineers, incorporating hardware constraints (latency, resource limits, numerical precision) as primary design considerations
• Translate model requirements into concrete decisions that shape the custom hardware roadmap
• Work closely with hardware engineers to implement, validate, and deploy ML inference solutions from prototype to production
• Evaluate emerging research (e.g., neural architecture search, ML systems, quantization) and apply what drives measurable system improvements

Your Skills and Experience

• Strong understanding of hardware constraints and trade-offs (e.g., pipelining, resource utilization, fixed-point arithmetic) for mapping ML models to FPGAs or ASICs
• Experience with hardware development (VHDL/SystemVerilog, HLS tools, or frameworks such as hls4ml, FINN, or Vitis AI)
• Solid grounding in machine learning fundamentals, including neural network architectures, inference optimization, and quantization techniques
• Proficiency in Python, C , or similar languages for tooling and simulation
• Ability to collaborate effectively across disciplines and communicate with both technical and non-technical stakeholders

Nice to Have

• Exposure to ML compiler stacks such as MLIR, TVM, or XLA
• Experience in latency-sensitive or resource-constrained environments (e.g., high-frequency trading, real-time systems, signal processing, particle physics)
• Familiarity with verification methodologies (SystemVerilog, UVM, Cocotb)
• Advanced degree in EE, CS, Physics, or a related field, or equivalent experience