About the Role
Join a stealth healthtech startup founded by an ex-Google engineer, building AI tools that empower clinicians to make faster, evidence-based decisions at the bedside.
We’re developing a cutting-edge platform that combines LLMs, retrieval-augmented generation (RAG), and vector search infrastructure to deliver real-time clinical insights. As the Founding Backend Engineer, you’ll play a critical role in shaping our backend systems, architecture, and engineering culture from the ground up.
What You’ll Do
* Build scalable backend microservices in Python (FastAPI) to support RAG workflows and user queries
* Develop and optimise vector search pipelines using tools like PGVector, Pinecone, or Weaviate
* Design embedding orchestration and hybrid retrieval mechanisms
* Implement evaluation frameworks (BLEU, ROUGE, hallucination checks) to monitor answer quality
* Deploy production systems on GCP (Cloud Run, Vertex AI, BigQuery, Pub/Sub)
* Own observability, IaC (Terraform), and CI/CD (GitHub Actions) pipelines
* Collaborate with product, mobile, and clinical experts to ship weekly improvements
* Ensure compliance with data privacy standards (GDPR, NHS DSPT)
Who You Are
* 5+ years of backend experience (ideally with Python)
* Proven track record delivering ML, search, or recommender systems at scale
* Deep knowledge of embeddings, LLM-based retrieval, and vector similarity search
* Hands-on with GCP (or AWS/Azure), Terraform, CI/CD, and observability
* Strong communicator, product-minded, and thrives in fast-paced startup environments
* UK-based and available to work 2–3 days per week in-office (London)
Bonus Points
* Experience in healthcare, medtech, or clinical systems
* Familiarity with MLOps tooling (MLflow, Kubeflow, Vertex Pipelines)