Research Engineer - Lab Brains & Scientific Workflows (12 months FTC)

Based at our London office, we're looking for a Research Engineer - Lab Brains & Scientific Workflows. Help build a research institute that remembers what it has learned and puts that knowledge to work. You can learn more about our ambitions here.

You will build Lab Brains: agents and connected tools that help laboratories understand their work, recover its history and follow through on what needs to happen next. You will work directly with researchers, technicians and lab managers, turning their recurring problems into software they choose to use.

Imagine asking why a protocol changed, receiving the relevant experimental record and meeting decision, and preparing the right follow-up in the lab's project system. Or helping a new researcher understand what has already been tried, which result remains uncertain and who can help. You will make workflows like these dependable, inspectable and easy to correct.

This is a hands-on technical leadership role with room to shape the product and influence the Institute’s use of AI. We want someone who thinks big and enjoys programming, trying unfamiliar tools, finding a small useful starting point and improving it through real use. You will be a founding member of the Institute-wide AI initiative, working as a peer with an engineer focused on scientific knowledge and field building.

What you'll build

• Useful lab workflows connecting meeting notes, approved Slack conversations, Linear projects and Labstep experimental records, with links to the relevant code, data and documents. Start with a small agreed pilot and extend what works.

• Agents and tools that recover experimental history, preserve decisions and negative or inconclusive results, identify missing context and propose follow-up that users can inspect and approve.

• Reusable integrations that handle changing records, permissions, retries and failures, and keep the authoritative systems connected without creating extra manual record-keeping.

• Interfaces and review controls that help people see where an answer came from, correct it and choose what may be shared beyond their lab.

• A tested route for distributing useful tools across participating labs, with documentation, dependable releases and clear ownership for support.

You will compare existing agents, scientific software and commercial tools on our actual workflows and develop a strategy for adoption across the Institute where they fit and leading development where a clear need remains. You will champion use of these agents to senior management, explaining the benefits they bring to the Institute and proactively revisit them as the ecosystem changes.

How you will work

Spend time with the people doing the science, then ship and improve small working products. Evaluate task completion, source accuracy, net time saved, recurring use, running costs and failure modes. Keep the mistakes and rejected suggestions visible so the team learns from them.

Treat reproducibility as part of the product: preserve the versions, methods, code and evidence needed to inspect a result or repeat an analysis. Build permissions, human approval and correction into the workflow. Researchers should gain useful memory and control over their work.

With the Scientific Knowledge & Field Building engineer, connect selected, authorised lab knowledge to wider scientific questions and return useful requests or opportunities to the lab. Both engineers share responsibility for an end-to-end workflow that improves real work.

What you bring

• Strong programming ability and evidence of shipping software that other people use, with the ability to learn the languages and tools needed by the codebase.

• Sound software engineering practice: Git with GitHub or equivalent, code review, automated testing, continuous integration, documentation, packaging, dependency management and maintainable releases.

• Practical experience building LLM applications or comparable intelligent automation, including tool/API integration, evaluation and debugging.

• Experience connecting systems and handling authentication, data access, changing records and operational failures.

• Product judgement: you can discover a useful problem with users, choose a manageable first version and improve it from feedback.

• A commitment to reproducible science and collaborative software development, demonstrated through examples of work others can inspect, run or improve.

• Enthusiasm and ability to influence how an organisation can drive change through technology adoption.

• The ability to work autonomously and lead: set priorities, make and justify technical and strategic decisions, bring others with you and take ownership of outcomes.

Experience with research software, laboratory workflows, scientific computing, developer tools, open-source packages, electronic lab notebooks or task systems would be valuable. You do not need prior experience with every named platform. We welcome evidence from industry, startups, academia, open source and independent projects; no particular degree or PhD is required where equivalent capability can be demonstrated.