AI Engineer - Agentic Systems

We're building the software infrastructure for the autonomous age of defence. AI is changing what software can do. The next step isn't just models that answer questions — it's systems that can reason, plan, use tools and take action. Our AI needs to operate within highly controlled, secure and mission-critical environments, which creates a completely different set of engineering challenges.

We're looking for an AI Engineer with strong software engineering fundamentals to help us build them.

Why you should join us

• You will build agentic systems that actually do things.
• We're moving beyond chatbots and simple LLM integrations.
• You will be building systems that can reason about a problem, plan a course of action, interact with tools, execute tasks and adapt based on what they discover.
• The interesting engineering challenge is making those systems reliable.
• How do you structure an agent for a particular task? When should it reason and when should it act? How do you give it access to the right tools without giving it too much freedom? How do you evaluate whether it's making the right decisions? How do you make the whole system secure, observable and predictable enough to operate in a high-stakes environment?

These are problems we're actively figuring out. You won't be taking an established recipe and implementing it. You'll have the opportunity to help define how agentic systems should actually be built for these environments.

This is an engineering role, not a prompt engineering role. We care deeply about AI, but we care just as much about software engineering. You will be designing architectures, building production services, integrating models into existing systems and thinking about everything required to make an AI system reliable in production.

You will work across the stack where necessary and have to think about things like:

• Architecture and system design
• Agent orchestration
• Tool use and Model Context
• Protocols
• LLM evaluation and reliability
• Backend services and APIs
• Data and state management
• Security and permissions
• Deployment in cloud and on-premise environments
• Observability and testing

The model is only one part of the system. We're looking for engineers who understand that.

You will solve problems that don't have an obvious answer. We're working with customers who have difficult problems and very high expectations. That means you'll need to be comfortable experimenting, throwing things away, trying different approaches and figuring out what works.

Sometimes the answer will be an LLM. Sometimes it will be a conventional algorithm or piece of software. Sometimes it'll be a combination of both. We're not interested in using AI for the sake of it. We're interested in building the best technical solution to the problem.

You will have genuine ownership. We're building out our AI capability at an exciting point in the company's growth. You will have significant autonomy from day one and the opportunity to influence how we approach agentic AI across the product.

You will work alongside strong engineers across software, AI and systems, with plenty of exposure to the wider technical architecture. If you have an idea for how something should work, we want you to be able to build it and prove it.

What we're looking for

• People who are strong software engineers first and genuinely excited about AI.
• A strong academic background in Computer Science, Mathematics, AI, Machine Learning, Engineering or a related technical field.
• Several years of hands-on software or ML engineering experience.
• Experience building agentic AI systems or LLM-powered applications in production.
• Strong programming ability in Python, Rust, C++, Java or another serious programming language.
• Strong understanding of software architecture and engineering fundamentals.
• Experience working with modern ML/AI frameworks and agent orchestration tools.
• An understanding of how LLMs work and the trade-offs involved in model selection.
• A pragmatic approach to solving difficult problems using both AI and conventional software engineering.
• Experience with frameworks such as LangGraph, Google ADK, MCP or similar agentic tooling is useful, but we don't expect you to have used our exact stack.
• Experience with production ML, MLOps, regulated environments or resource-constrained systems would also be valuable.

What you'll get

• Highly competitive salary
• Meaningful equity
• Private healthcare and dental
• Free meals in the office
• Relocation support
• Access to the compute, equipment, training and resources you need
• A highly technical, collaborative engineering environment
• The opportunity to work on some of the most interesting applications of AI being built today

We're based in London and work together in the office by default. We're a small team, we're moving quickly and we think there's a huge amount of value in building alongside each other. There will also be opportunities to work directly with customers and see the systems you're building being used in real environments.

If you're a strong software engineer who wants to work on the next generation of AI systems — not just wrap an API around an LLM — this is an opportunity to get involved at the ground floor. We're trying to figure out what agentic AI looks like when reliability, security and real-world impact actually matter. If that sounds like the kind of engineering problem you want to spend your time solving, we'd like to hear from you.