Senior Software Engineer, AI Platform
Job Description
Color’s AI Platform team builds full-stack AI products and the platform that powers them across Color. This onsite role in San Francisco, CA focuses on reliable LLM-powered systems, evaluation and safety practices, and privacy- and security-conscious healthcare solutions.
Compensation: USD 175,000 to 230,000 per year. Experience level: 5+ years.
Benefits
- Competitive salary
- Comprehensive medical, dental, vision, life, and disability benefits
- 401k match
- Monthly phone and wifi stipend for employees, plus an annual ergonomic stipend
- Generous vacation policy, paid holidays, and company-wide recharge days
- Equal paid parental leave for birthing and non-birthing parents
- Free cancer screening and prevention resources for employees and their adult dependents
Responsibilities
- Build full-stack AI applications for Color’s care teams and patients, owning features from user experience through backend services, AI orchestration, data access, and production deployment
- Explore how emerging AI capabilities can improve cancer care and how Color works, moving from ideas to prototypes to shipped experiences
- Experiment quickly with new AI approaches, turning promising models, tools, and techniques into prototypes and evolving them into production systems
- Develop reliable LLM-powered systems including retrieval and knowledge systems, citations and grounding, prompt management, structured outputs, tool use, agentic workflows, and evaluation
- Create the evaluation, observability, and safety practices needed to understand quality, compare approaches, identify regressions, and confidently evolve successful experiments into durable products
- Design privacy- and security-conscious systems for sensitive healthcare data and clinical workflows
- Partner with product, design, clinical, and engineering teams to understand problems, prototype solutions, learn from users, and ship improvements quickly
- Increase Color’s AI fluency by sharing experiments, emerging techniques, lessons learned, and strong patterns for building with AI
- Mentor others and help raise the technical and product quality bar
Requirements
- 5+ years building and shipping production software, with strong full-stack skills across React/TypeScript, Python, APIs, data integrations, and cloud infrastructure
- Hands-on experience building production AI products with modern LLMs, including patterns such as RAG, structured outputs, tool calling, agentic workflows, prompt engineering, and model evaluation
- Ability to turn complex, heterogeneous data into useful product experiences and move systems from prototype to production, including testing, monitoring, and iteration
- Strong product judgment to balance speed, technical quality, reliability, and user impact in ambiguous, rapidly changing environments
- A bias toward tinkering and rapid experimentation, including the ability to start small, prototype quickly, measure results, and recognize when a new capability enables a larger leap
- An informed point of view on how AI is changing software engineering and product development, including regular experimentation with models, frameworks, coding tools, agent architectures, and interaction patterns, plus communicating what you learn
Technologies
React, TypeScript, Python, APIs, RAG, cloud infrastructure
Possible Projects to Contribute To
- Healthbook: Build and evolve an internal AI application that helps care and clinical teams understand patients’ longitudinal health histories, synthesize records, answer grounded questions with citations, and draft documents using an AI agent
- AI Platform: Build shared capabilities for teams across Color to develop reliable AI applications quickly, including knowledge bases and retrieval, citations and grounding, prompt management, evaluation, observability, and reusable agentic patterns
- What’s next: Explore next-generation patient-facing AI experiences, then turn learnings into platform capabilities teams can use to build with AI
Nice to Have
- Experience building AI platforms or shared infrastructure that helped multiple teams ship AI applications
- Experience developing healthcare technology involving sensitive clinical data