Senior AI Engineer

We are working with a fast-growing UK frontier technology company developing advanced AI systems for organisations operating in complex, high-stakes environments., The business is looking for a Senior Applied AI Engineer to take genuine technical ownership of the design and delivery of production-grade agentic AI systems.

This is a builder’s role rather than an integration role. You will design and implement sophisticated AI systems from the ground up, make architecture-level decisions, write production-quality software and own features through to deployment.

The focus is on turning rapidly evolving AI capabilities into reliable, scalable systems that deliver measurable value in real-world commercial environments.

You will work across agent orchestration, LLM and vision-language model integration, retrieval, multimodal data, evaluation and production infrastructure. You’ll be expected to understand not only how to make an AI system work, but how to make it dependable, observable and maintainable at scale. What You’ll Be Doing

• Design and build multi-agent AI systems from the ground up, using frameworks such as LangGraph, LangChain, Haystack or similar where appropriate.
• Develop orchestration, state management and tool-calling infrastructure required to make agentic systems reliable in production.
• Integrate large language models and vision-language models into reasoning, retrieval, summarisation and task-execution workflows.
• Build retrieval, memory, evaluation and guardrail capabilities around AI systems.
• Design and implement production pipelines covering data ingestion, processing and inference.
• Build search and retrieval systems across very large multimodal datasets, including text, imagery, telemetry and sensor data.
• Design indexing, embedding and querying infrastructure capable of operating across multi-terabyte datasets.
• Engineer systems for performance, reliability, scalability and predictable failure behaviour.
• Deploy AI systems across cloud and on-premises environments, with an understanding of the constraints associated with each.
• Build evaluation and observability capabilities to measure model performance, agent behaviour and system reliability.
• Take end-to-end ownership of technical workstreams, from architecture and implementation through to deployment.
• Make systems-level design decisions and establish engineering patterns for other developers to follow.
• Produce clear technical documentation and knowledge transfer to ensure the systems you build can be maintained and extended by the wider team., This isn’t a role focused on experimenting with AI in notebooks or stitching together a collection of third-party APIs.

You’ll be working on production systems where the engineering around the models matters just as much as the models themselves.

That means thinking carefully about:

• How agents reason and interact with one another.
• How context and memory are managed.
• How information is retrieved from very large datasets.
• How models behave under real workloads.
• How latency and reliability are controlled.
• How failures are detected and handled.
• How system behaviour is evaluated and monitored.
• How AI systems can be deployed securely and maintained over time.

You will have significant autonomy and will be expected to contribute to the technical direction of the wider platform. Why This Role?

This is an opportunity to work on genuinely challenging applied AI problems and see your work move from architecture through to production.

You’ll ideally have:

• 6+ years of professional software engineering experience, including several years working with LLM-based, generative AI or agentic systems.
• A demonstrable track record operating at Senior, Lead or equivalent engineering level.
• Commercial experience building and shipping multi-agent or agentic AI systems into production.
• Strong Python development skills and excellent software engineering fundamentals.
• Hands-on experience with LangGraph, LangChain, Haystack or similar AI/LLM frameworks, together with the ability to work below the framework level when necessary.
• Experience designing and deploying AI/ML systems into genuine production environments.
• Strong understanding of model inference, latency, performance, data pipelines, state, memory and tool/function calling.
• Hands-on experience building search and retrieval systems over very large multimodal datasets.
• Experience with Docker, Git and cloud platforms, ideally AWS.
• An understanding of distributed systems, production infrastructure and scalable application design.
• The ability to take technical ownership and make sound architecture-level decisions.
• A pragmatic engineering mindset: you care about whether a system works reliably in production, not just whether a prototype looks impressive.
• Strong communication skills and the ability to explain complex technical concepts clearly to both technical and non-technical stakeholders.

Desirable Experience

The following would be advantageous:

• Multimodal AI and reasoning.
• Edge or offline AI deployments.
• Kubernetes, particularly EKS or OpenShift.
• MLOps, including model evaluation, monitoring and reproducibility.
• Observability for agentic AI systems, including model performance, agent behaviour and drift.
• Agent orchestration and inter-agent communication protocols such as A2A.
• Model Context Protocol (MCP).
• Secure-by-design development principles, including ISO 27001, NIST or OWASP.
• Experience working with highly regulated, mission-critical or data-intensive organisations.
• Experience with large-scale data platforms and distributed search.
• Contributions to open-source AI/ML projects., If you’re a senior software engineer who enjoys building sophisticated AI systems from first principles - and you’re interested in taking ownership of production-grade agentic AI rather than simply experimenting with the latest models - we’d like to hear from you.

Please apply with an up-to-date CV highlighting your experience with Python, LLMs, agentic AI, production software engineering, retrieval systems and cloud infrastructure.

• Own significant technical workstreams end-to-end.
• Shape architecture rather than simply deliver predefined tickets.
• Work with large-scale multimodal datasets.
• Solve difficult problems around AI reliability, retrieval, orchestration and deployment.
• Work alongside a highly technical, fast-growing team.
• See your engineering directly influence products used by real organisations.
• Potentially extend the engagement into a longer-term opportunity.

What We Offer

• £700-£900 per day, depending on experience.
• Outside IR35 engagement.
• Initial 6-month contract with a strong likelihood of extension.
• Genuine technical ownership and influence over architecture.
• Hybrid working: 2 days per week on site in Harwell, Oxfordshire, with the remainder remote.
• Immediate start.
• Opportunity to work at the forefront of agentic and generative AI.
• Potential for a longer-term engagement with a rapidly growing technology business.

Working Pattern

This is a UK-based hybrid contract position.

You will be expected to work 2 days per week on site in Harwell, Oxfordshire, with the remaining time worked remotely.