Location
London

Hours
Full Time

Salary
Competitive, commensurate with experience

About the Role
Unilabs, headquartered in Geneva and part of the A.P. Møller Group, is one of Europe’s leading medical diagnostics companies, providing laboratory, pathology, genetics, and imaging services across 14 countries. We invest heavily in technology and people to improve the lives of nearly 100 million patients annually through state-of-the-art digital laboratories and imaging institutes.

We are seeking an AI Native Engineer to tackle complex, high-impact challenges using large language models (LLMs), agentic frameworks, and modern AI tooling. In this role, you will design and deploy production-grade AI systems that transform millions of unstructured pathology and genomics records into actionable clinical insights, directly influencing healthcare innovation and patient outcomes at scale.

Core Responsibilities
1. Core Agentic Architecture & Retrospective Extraction: Design, build, and maintain LLM-based extraction pipelines to parse years of unstructured PDF pathology reports. Extract clinical entities such as diagnoses, tumor grades, pathological staging, and biomarker statuses from free-text documents. Evaluate and integrate agentic frameworks (e.g., LangChain, LlamaIndex) based on extraction accuracy. Develop confidence scoring systems and human-in-the-loop validation queues for clinical review.

2. Multi-Modal Pipeline & Next-Gen API Infrastructure: Architect data pipelines linking pathology LIS data with molecular/genetics systems to ensure seamless mapping of vital markers to case records. Implement REST APIs, HL7 v2, or HL7 FHIR interfaces for downstream integration with platforms like Proscia Concentriq and Aperture. Develop secure, high-throughput API layers for future integration with consumer wearables, preventive health apps, and cloud-native hospital systems. Build automated data-quality monitoring systems to detect anomalies and incomplete records.

3. Governance, De-Identification & Compliance: Implement de-identification protocols to pseudonymize patient identifiers. Ensure full compliance with global and regional health data privacy regulations including Swiss nDSG and EU GDPR Article 9. Build audit logging and data lineage tracking to preserve clinical data provenance for pharma and clinical partners.

Experience
- Practical, hands-on experience with LLM APIs, prompt engineering, and building system prompt state machines for structured text extraction.
- Direct experience with agentic frameworks such as LangChain or LlamaIndex to orchestrate complex clinical data workflows.
- Proven track record of shipping non-deterministic AI models into production, managing context windows, token costs, rate limits, and output evaluation.
- 4–7+ years of software engineering experience with strong proficiency in Python and SQL, capable of independently debugging asynchronous pipelines.
- Experience designing and consuming production-grade REST APIs in regulated or clinical environments.
- Practical deployment experience on cloud platforms (AWS, Azure, or GCP) using Docker containerization.
- Preferred: Knowledge of clinical data standards such as HL7 v2, FHIR, OMOP CDM, and CDISC.
- A plus: Exposure to digital pathology data formats (DICOM, SVS, NDPI) or LIS systems.

About You
You are passionate about applying AI to healthcare challenges, detail-oriented, and thrive in a fast-paced, innovative environment. You understand the critical importance of data privacy and compliance in clinical settings and are committed to delivering robust, scalable AI solutions that improve patient outcomes.

Qualifications
Relevant degree or equivalent experience in computer science, software engineering, bioinformatics, or related fields is expected. Strong communication skills and the ability to collaborate with clinical and technical teams are essential.