Key skills
Agentic AI systemsLarge Language Models (LLMs)Retrieval-Augmented Generation (RAG) architecturesVector storesLLM evaluationPythonPandasPyTorchTerraformAWSDockerCI/CD deployment pipelinesTesting frameworksTeam player attitudeAILLMRAGAgenticCI/CD
About this role
23andMe is looking for a Senior Software Engineer to help build the AI systems that turn one of the world's largest genetic datasets into personalized health insights. You will design and operate agentic AI pipelines and ensure the quality of AI-generated results for health products.
Responsibilities:
- Build agentic AI systems that generate personalized health and ancestry insights for customers, orchestrating LLMs with relevant genetic and clinical context
- Design and implement RAG pipelines that retrieve relevant context from vector stores to ground model outputs in a customer's specific genetic data
- Build robust evaluation frameworks to measure model quality, catch regressions, and ensure AI-generated results meet the bar for a health product
- Build and rely on CI/CD methods and automatic testing to ensure reliable code, continuously delivered
- Debug production incidents and get services back online – you'll own what you ship
- Integrate AI into daily workflows to automate tasks and improve efficiency
Requirements:
- B.S. in Computer Science or equivalent; 4+ years of industry experience
- Hands-on experience building agentic AI systems and LLM-powered pipelines in production
- Experience with RAG architectures and vector stores
- Strong instincts around LLM evaluation: you understand why 'it looks good' isn't enough for a health product
- Proficiency with Python and related scientific libraries (pandas, pytorch)
- Infrastructure as code experience, ideally in Terraform on AWS, containerized in Docker
- Hands-on experience working with CI/CD deployment pipelines and testing frameworks
- A team player attitude with comfortability operating cross-functionally across the organization