Skip to main content

Senior Machine Learning Systems Engineer, LLM Inference Performance on AWS Trainium

Summary by AIEngineer.careers

Senior Machine Learning Systems Engineer, LLM Inference Performance on AWS Trainium is a ML Systems Engineer role (full-time). with Capgemini. in REMOTE, TORONTO, CA. Compensation shown: $48K–$93K (est.). Imported listing (source: movojobs.ca). Apply on the employer's site (movojobs.ca).

Imported listing (source: movojobs.ca) · Apply on movojobs.ca

How our listings work

REMOTE · TORONTO · CA
$48K–$93K (est.)
Full-time
SENIOR
Apply Now

More AI jobs in Canada · Remote AI jobs

The sections below reproduce the third-party job description for reference. AIEngineer.careers does not write or control this text.

Imported job description

Sourced from movojobs.ca

Capgemini Engineering is hiring a hands-on ML systems engineer to optimize production-grade LLM inference on AWS Trainium platforms (Trainium2/Trainium3). The role focuses on decode latency, throughput, price-performance, long-context and multi-chip scaling, and reproducible benchmarking against leading GPU platforms. You will work with vLLM on AWS Neuron, speculative decoding, and multi-token prediction for Qwen-family models. Remote anywhere in Canada.

Responsibilities

  • Improve decode latency, throughput, and cost on AWS Trainium platforms while validating correctness
  • Implement speculative decoding and native multi-token prediction (MTP) for Qwen-family models in vLLM on AWS Neuron
  • Design and run matched, reproducible benchmarks against leading GPU platforms
  • Deliver latency-throughput Pareto frontiers and price-performance comparisons aligned with customer SLOs
  • Benchmark and migrate workloads across Trainium generations
  • Scale serving on multi-chip Trainium systems for long-context workloads (100K+ tokens)
  • Onboard additional model architectures including hybrid state-space and mixture-of-experts models
  • Create runbooks, implementation guides, and benchmark documentation; support customer workshops

Requirements

  • 5+ years in machine learning or systems engineering, including 2+ years on production model inference or serving
  • Strong proficiency in Python and PyTorch, able to modify model implementations involving attention, MoE routing, and KV-cache management
  • Hands-on experience with an LLM serving framework such as vLLM, TensorRT-LLM, SGLang, or Triton Inference Server
  • Deep understanding of LLM inference performance: prefill/decode, continuous batching, tensor and expert parallelism
  • Experience profiling and optimizing accelerator workloads at framework, compiler, or kernel level
  • Strong written and verbal communication skills

Nice to Have

  • Experience with AWS Trainium/Inferentia and the AWS Neuron SDK
  • Accelerator migration experience
  • Speculative decoding techniques
  • Kernel development using NKI, Triton, or CUDA

Benefits

paid time offHealth insurancedental insurancevision insuranceretirement planlife insuranceemployee assistance program

Tech Stack

AWS Neuron SDKAWS TrainiumCUDAKV-cacheLLM InferenceMixture of ExpertsNKIPythonPyTorchSGLangspeculative decodingTensorRT-LLMTriton Inference ServervLLM
Capgemini

Capgemini

Paris · FR · 348366+ employees

Capgemini is an AI-powered global business and technology transformation partner with nearly 60 years of heritage, operating in more than 50 countries. The publicly listed group (CAP.PA) reported 2025 global revenues of €22.5 billion and delivers end-to-end services across strategy, technology, design, engineering, and business operations.

Weekly job alert

Get a weekly email when new AI engineering jobs match your filters. Free — confirm your address to start.

Related Jobs