Client-Facing AI Model Engineer (AWS)

Aws Emea Sarl (Uk Branch); England; On-site; GBP 110,000 - 150,000; Full time

Job summary

Aws Emea Sarl (Uk Branch) is hiring a Forward Deployed Engineer II to embed with customer teams, defining evaluation bars and technical plans for model workstreams. You will customize open-weight models, perform post-training tuning, and build scalable training and inference pipelines on AWS Trainium.

You will optimize deployments, improve latency and cost, and share reusable recipes with AWS service teams. Travel up to 25% is required as you engage with customers and engineers.

Qualifications

• Professional software development experience (non-internship).
• Proficiency in at least one modern language (Java/C++/C#) and OO design.
• 3+ years of non-internship software dev experience.
• 2+ years in Agile, startup, or fast-scaling environment.
• Hands-on experience with generative AI technologies (LLMs, vision-language, speech).
• Experience with AI tools: eval frameworks, agent tooling, RAG pipelines, prompt engineering.
• Ability to communicate technical concepts to diverse audiences.

Responsibilities

• Embed with customer ML and engineering teams; translate goals into a measurable plan.
• Customize and post-train models; perform continued pre-training and RLHF.
• Build distributed training and inference pipelines on AWS Trainium.
• Optimize models on AWS chips; improve throughput, latency and cost.
• Codify reusable fine-tuning recipes and share learnings with AWS teams.

Skills

Software development

OO programming

3+ years experience

Agile environment

Generative AI

AI tools

Communication skills

Education

Bachelor's degree in CS or equivalent

Tools

PyTorch FSDP2

Megatron-LM

DeepSpeed

NeMo

TorchTitan

Miles

verl

Job description

Aws Emea Sarl (Uk Branch) is hiring a Forward Deployed Engineer II to embed with customer teams, defining evaluation bars and technical plans for model workstreams. You will customize open-weight models, perform post-training tuning, and build scalable training and inference pipelines on AWS Trainium.

You will optimize deployments, improve latency and cost, and share reusable recipes with AWS service teams. Travel up to 25% is required as you engage with customers and engineers.