AI Engineer — Clinical Documentation

Hana Novak

Full-time · Mid-level · London

This role is no longer accepting applications

Similar open roles you can apply to now:

About the role

We build software for hospital systems. Specifically, we build a platform that helps clinical teams produce, structure, and manage patient documentation — discharge summaries, care plans, referral letters, and clinical notes. We have 38 NHS trust customers in the UK and are expanding into the US market this year. Our AI team has spent the last eighteen months embedding LLM-powered features into the documentation workflow: draft generation from structured encounter data, automated coding suggestions against ICD-10 and SNOMED CT, and a semantic search layer over the patient record. These features are in production. Clinicians use them every day. The feedback loop is immediate and honest — when the AI produces something wrong or confusing in a clinical context, a consultant will tell us about it within hours. We are hiring an AI Engineer to join the team building these features. The work is technically interesting: the domain vocabulary is highly specialised, the documents follow strict structural conventions, and the consequences of hallucination are not abstract. We expect candidates to take data governance and patient safety seriously — not as compliance overhead, but as a fundamental constraint that shapes good engineering in this domain.

Responsibilities

Build and improve LLM-powered draft generation for clinical document types: discharge summaries, care plans, and referral letters
Develop and maintain the semantic search layer over structured patient records
Design evaluation frameworks for clinical AI outputs — hallucination detection, factual consistency, and clinical appropriateness scoring
Work directly with NHS clinical informatics teams to understand documentation workflows and translate requirements into engineering specifications
Contribute to our data governance documentation and support clinical safety reviews of AI features

Requirements

4+ years of software or ML engineering with at least 2 years building LLM-powered features used by real end users
Python — clean, production-quality code that clinical engineers can read and audit
LangChain or LlamaIndex for RAG and structured generation workflows in domain-specific document environments
OpenAI or Anthropic API at depth: structured outputs, function calling, prompt design, and cost management at scale
RAG pipeline design with particular attention to retrieval precision in specialised vocabulary contexts
SQL for working with patient encounter data and understanding how structured clinical data feeds generation pipelines
Familiarity with GDPR, NHS DSPT, or HIPAA data handling requirements — you understand why these exist and how they affect system design

Benefits

$88,000 – $112,000 base salary
Full remote — our engineering team is distributed across the UK
Work on a problem where the feedback from real clinicians is immediate and the quality bar is set by patient safety, not metrics
$1,500 annual learning budget
25 days annual leave plus bank holidays