A senior machine learning systems engineer will optimize LLM inference performance with Python, PyTorch, and AWS Trainium remotely anywhere in Canada. The focus includes decode latency and throughput.
At Capgemini Engineering, this hands-on engineer improves production-grade model serving across Trainium2 and Trainium3. The work covers vLLM on AWS Neuron, Qwen-family models, speculative decoding, multi-token prediction, and reproducible benchmarking against GPUs. You will address long-context and multi-chip scaling.

Key Responsibilities

Improve decode latency, throughput, cost, and correctness on TrainiumImplement speculative decoding and native multi-token predictionDesign matched tests controlling prompts, concurrency, cache,



and warm-upScale long-context serving across multi-chip Trainium systemsCreate runbooks, guides, documentation, and hands-on workshopsRequirements:
Five-plus years in machine learning or systems engineeringTwo-plus years in production model inference or servingStrong Python and PyTorch proficiency modifying model implementationsExperience with vLLM, TensorRT-LLM, SGLang, or TritonUnderstanding of prefill, decode, batching, parallelism, and costApply machine learning systems expertise to optimize AWS Trainium inference, benchmarking, and serving performance at Capgemini Engineering.
#J-18808-Ljbffr

📌 Senior Llm Inference Performance Engineer (Winnipeg)
🏢 Capgemini Engineering
📍 Winnipeg