Machine Learning Engineer - GPU Performance & Low-Latency Inference
Pay · estimate~$145k – $195kMarket rate · not set by employer
PostedNov 17, 2026
LevelMid level
In short
We’re partnering with a leading global trading firm to hire a Machine Learning Engineer to make ML models run as fast as possible.
📋 About the job
Job Description
We’re partnering with a leading global trading firm to hire a Machine Learning Engineer to make ML models run as fast as possible. You’ll work close to the hardware, squeezing latency out of real-time inference and speeding up training, in an environment where microseconds matter.
\n
\nWhat you’ll do
\n\n
- Build and optimize low-latency inference pipelines that deliver real-time predictions in production
\n
- Profile and optimize ML workloads at the device level: memory, compute, kernels and compilers
\n
- Write and tune custom GPU code (CUDA, Triton or equivalent) where libraries fall short
\n
- Apply and validate low-precision techniques (FP8/INT8, quantization), knowing which optimizations stay bit-exact,
\n
- Optimize batching, KV caching and serving paths for latency-critical workloads.
\n
- Extend the internals of PyTorch and inference frameworks to raise performance on specific workloads
\n\n
\nWhat you’ll bring
\n\n
- 3+ years in machine learning, focused on inference or GPU performance
\n
- Hands-on experience optimizing ML at the GPU level: CUDA, Triton, C++, TensorRT or similar
\n
- Experience with profiling tools (Nsight, torch.profiler) and reasoning about memory bandwidth, latency and throughput
\n
- A strong grasp of modern ML architectures, attention in particular, and how they map onto hardware
\n
- Production-quality Python (C++ a strong plus), with testing discipline and ownership of the systems you ship
\n
- A habit of measuring before optimizing, and the judgment to know when to build versus buy
\n\n
\nNice to have
\n\n
- Experience building real-time, low-latency prediction systems in high-performance environments
\n
- Experience with numerics: mixed precision, quantization and correctness at scale
\n
- Contributions to open-source projects such as PyTorch, vLLM, Triton or FlashAttention
\n\n
\nNo trading experience is required. What matters is deep, hands-on performance engineering and a drive to make things faster.
Interested in this role?Apply in one click — no JobQ account needed.
Apply now →