Machine Learning Engineer, Scarlet

• Creator of datasets – You roll up your sleeves and build the dataset you wish existed, not wait for someone else to give you one
• Establisher of evals – You think from first principles about statistics and how to evaluate ML systems, and you’re suspicious of benchmarks that don’t match reality
• Tuner of models – ****Fine‑tuner, GPU memory leak fixer, agent swarm orchestrator
• Shipper of software – 3+ years experience shipping software to production
• Hacker – Track record of taking ambitious projects from concept to reality with whatever tools are at hand
• Builder – You’re insatiably curious about real‑world problems and care that what you build has clear economic and human value

Desirable

• The Applied Machine Learning team at Scarlet works across the entire business to develop production tools that superpower our customers, and our internal teams, to achieve certification orders of magnitude faster and more reliably than previously possible
• The Applied Machine Learning team works closely with medical doctors, AI researchers, and engineers to identify, prototype, validate, deploy and monitor AI systems that are fundamental to Scarlet’s mission
• As a Machine Learning Engineer in Scarlet’s Applied Machine Learning team, you will have the opportunity to work with some of the brightest minds in AI medical devices
• You’ll deploy state-of-the-art models in production environments, evaluate performance and iterate
• If you're excited about building distributed agentic systems, fine tuning LLMs, and creating exceptional datasets, this role is your chance to make a significant impact on the quality of medical devices that go to market
• You might, for example, work on:
• Agentic document search Build concurrent agent architectures to for rapid, multi‑agent evidence gathering across messy, distributed customer data sources. Deploy to production
• Evals wired into real workflows Build evals that use our production assessment workflows to gather ground truth: IoU matching, precision/recall, agreement with human judgements, and other metrics that actually move the needle for certification timelines
• Applied AI alignment Build customer support agents that delight customers and conform to our strict impartiality and objectivity commitments as a regulated certification agency

• Intro call with Alan (team lead) – mutual sanity check, space for your questions
• Working session in our London office with Alan, Jamie (CTO), and team – walk through something you’ve built that you’re proud of; expect us to poke at your decisions and trade‑offs. Then riff with us on a real problem we're working on