Scarlet

Machine Learning Engineer

• Build agentic document-understanding systems that search, parse and visually inspect technical files
• Develop agent harnesses to retrieve information across large document collections in varied formats
• Preserve source attribution and minimise hallucination while balancing accuracy, latency and cost
• Deploy agent systems to production
• Define success with assessors and build datasets and benchmarks for a complex domain
• Measure retrieval quality, citation correctness, expert agreement, assessor effort, assessment quality, customer experience and rework
• Use evaluation results to prioritise improvements
• Build agents that respect the impartiality and objectivity required of a certification body
• Help users understand evidence, recognise uncertainty and retain responsibility for consequential judgments
• Own ML systems from conception and prototyping through deployment, evaluation and iterative improvement
• Collaborate with clinicians, assessors and engineers to accelerate medical-device certification without compromising safety