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

At Capgemini Engineering, the world leader in engineering services, we bring together a global team of engineers, scientists, and architects to help the world’s most innovative companies unleash their potential. From autonomous cars to life-saving robots, our digital and software technology experts think outside the box as they provide unique R&D and engineering services across all industries. Join us for a career full of opportunities. Where you can make a difference. Where no two days are the same.

Location

Remote anywhere in Canada

About the job you’re considering

We are seeking a hands-on machine learning systems engineer to optimize production-grade large language model inference across current and next-generation AWS Trainium platforms, including Trainium2 and Trainium3. The work focuses on decode latency, throughput under representative load, price-performance, long-context and multi-chip scaling, and reproducible benchmarking against leading GPU platforms.

Your role

• Improve decode latency, throughput, and cost on AWS Trainium platforms, validating correctness at every step. Initial priorities include speculative decoding and native multi-token prediction (MTP) for Qwen-family models in vLLM on AWS Neuron.

• Design and run matched, reproducible tests against leading GPU platforms, controlling prompts, concurrency, cache state, and warm-up. Deliver latency-throughput Pareto frontiers and price-performance comparisons aligned with customer service-level objectives.

• Benchmark and migrate workloads across Trainium generations, including Trainium2 and Trainium3. Scale serving on multi-chip Trainium systems for long-context workloads, including contexts of 100K tokens or more, and apply architecture-specific optimizations.

• Extend model coverage. Onboard additional architectures, including hybrid state-space and mixture-of-experts models, reusing and generalizing existing implementations where practical.

• Create runbooks, implementation guides, and benchmark documentation, and support hands-on workshops so customer engineers can operate, reproduce, and tune the stack independently.

Your skills and experience

• 5+ years of experience in machine learning or systems engineering, including 2+ years focused on production model inference or serving.

• Strong proficiency in Python and PyTorch, with the ability to read and modify model implementations involving attention, mixture-of-experts 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, including prefill and decode, continuous batching, tensor and expert parallelism, and their impact on latency, throughput, and cost.

• Experience profiling and optimizing accelerator workloads at the framework, compiler, or kernel level, and designing rigorous performance experiments.

• Strong written and verbal communication skills, with the ability to explain engineering trade-offs to technical audiences and business impact to senior stakeholders.

• Preferred qualifications include experience with AWS Trainium/Inferentia, the AWS Neuron SDK, and optimizing AI inference workloads, along with expertise in accelerator migrations, speculative decoding techniques, and kernel development using NKI, Triton, or CUDA.

The base compensation range for this role in the posted location is: $72,056 - $138,940/year

Capgemini provides compensation range information in accordance with applicable national, state, provincial, and local pay transparency laws. The base compensation range listed for this position reflects the minimum and maximum target compensation Capgemini, in good faith, believes it may pay for the role at the time of this posting. This range may be subject to change as permitted by law.

The actual compensation offered to any candidate may fall outside of the posted range and will be determined based on multiple factors legally permitted in the applicable jurisdiction.

These may include, but are not limited to: Geographic location, Education and qualifications, Certifications and licenses, Relevant experience and skills, Seniority and performance, Market and business consideration, Internal pay equity.

It is not typical for candidates to be hired at or near the top of the posted compensation range.

In addition to base salary, this role may be eligible for additional compensation such as variable incentives, bonuses, or commissions, depending on the position and applicable laws.

Capgemini offers a comprehensive, non-negotiable benefits package to all regular, full-time employees. In the U.S. and Canada, available benefits are determined by local policy and eligibility and may include:

· Paid time off based on employee grade (A-F), defined by policy: Vacation: 12-25 days, depending on grade, Company paid holidays, Personal Days, Sick Leave

· Medical, dental, and vision coverage (or provincial healthcare coordination in Canada)

· Retirement savings plans (e.g., 401(k) in the U.S., RRSP in Canada)

· Life and disability insurance

· Employee assistance programs

· Other benefits as provided by local policy and eligibility

Important Notice: Compensation (including bonuses, commissions, or other forms of incentive pay) is not considered earned, vested, or payable until it becomes due under the terms of applicable plans or agreements and is subject to Capgemini’s discretion, consistent with applicable laws. The Company reserves the right to amend or withdraw compensation programs at any time, within the limits of applicable legislation.

Disclaimers

Capgemini is an Equal Opportunity Employer encouraging inclusion in the workplace. Capgemini also participates in the Partnership Accreditation in Indigenous Relations (PAIR) program which supports meaningful engagement with Indigenous communities across Canada by promoting fairness, accessibility, inclusion and respect. We value the rich cultural heritage and contributions of Indigenous Peoples and actively work to create a welcoming and respectful environment. All qualified applicants will receive consideration for employment without regard to race, national origin, gender identity/expression, age, religion, disability, sexual orientation, genetics, veteran status, marital status or any other characteristic protected by law.

This is a general description of the Duties, Responsibilities and Qualifications required for this position. Physical, mental, sensory or environmental demands may be referenced in an attempt to communicate the manner in which this position traditionally is performed. Whenever necessary to provide individuals with disabilities an equal employment opportunity, Capgemini will consider reasonable accommodations that might involve varying job requirements and/or changing the way this job is performed, provided that such accommodation does not pose an undue hardship. Capgemini is committed to providing reasonable accommodation during our recruitment process. If you need assistance or accommodation, please reach out to your recruiting contact.

Please be aware that Capgemini may capture your image (video or screenshot) during the interview process and that image may be used for verification, including during the hiring and onboarding process.

Click the following link for more information on your rights as an Applicant in the United States.

Capgemini is a global business and technology transformation partner, helping organizations to accelerate their dual transition to a digital and sustainable world, while creating tangible impact for enterprises and society. It is a responsible and diverse group of 340,000 team members in more than 50 countries. With its strong over 55-year heritage, Capgemini is trusted by its clients to unlock the value of technology to address the entire breadth of their business needs. It delivers end-to-end services and solutions leveraging strengths from strategy and design to engineering, all fueled by its market leading capabilities in AI, generative AI, cloud and data, combined with its deep industry expertise and partner ecosystem.