Senior Forward Deployed Engineer (AI)

About Cressoft

Cressoft is a UK-based IT consultancy, operating since 2011 and delivering elite technical talent to global brands. We’re proud to be the exclusive offshore partner supporting a major engineering-led digital transformation for one of the UK’s largest supermarket chains.

• UK based candidates only. It is an onshore role, hybrid, up to 2 days a week in the office
• Contract: 12 months initially
• Start: as soon as possible, ideally November 2026

Senior AI Forward Deployed Engineer (Contract)

The role

A large UK retailer is looking for a Senior AI Forward Deployed Engineer to work directly with its product and business teams and turn their problems into working AI systems in production.

You will sit close to the people who own the problem, build the solution yourself on large language models, ship it, and stay with it until it delivers a measurable result. The work includes AI-driven search and conversational shopping products that serve millions of customers and must hold up at peak trading. This is a hands-on engineering role. It is not a research role and not an architecture-only role.

What you will do

• Work directly with business and product teams to find where AI will make a real difference, and scope it into something that can ship
• Build it yourself: prototype quickly, prove the value, then harden it for production
• Own what you ship end to end: retrieval, agents, prompts and context, streaming responses, and live product data such as price and stock
• Productionise properly: deployment, monitoring, cost control and reliability at high traffic
• Set up evaluation so quality is measured, not guessed: test sets, automated checks and monitoring in production
• Make sound trade-offs on latency, cost and accuracy, and explain them in plain terms
• Raise the level of the team: review designs and code, coach engineers, and help interview new hires
• Keep the team current on a fast-moving field and decide what is worth adopting

What you will bring

• Around 10 years in software engineering, with a strong backend background. Some frontend knowledge is enough
• A record of shipping LLM-based products to production yourself: retrieval augmented generation (RAG), agents and prompt-based applications
• Depth in context engineering and in evaluations
• Solid fundamentals: how LLMs and embedding models work, vector databases, context windows and their limits
• Strong system design: you can design a system from scratch, make clear decisions and defend them
• Experience running AI systems in the cloud at scale, with operational ownership after launch
• Experience working face to face with non-technical stakeholders, turning a loosely defined problem into a delivered system

Not what we are looking for

• A background centred on model training, PyTorch, transformer internals, classical machine learning or GPU work, without the application engineering above
• Architects or advisers who design but no longer build

Nice to have

• Search experience: relevance, ranking, hybrid or semantic search. This is a big plus
• E-commerce or retail experience