Senior Software Engineer, Agentic AI

About the company Heartflow is a publicly traded medical technology company (HTFL) using AI to advance the diagnosis and management of coronary artery disease — the #1 cause of death worldwide. Their flagship product generates a personalized 3D model of a patient's coronary arteries from a single CT scan, cleared for use in the US, UK, Europe, Japan, and Canada. To date, Heartflow has helped clinicians manage care for nearly 650,000 patients across more than 1,800 institutions worldwide. About the role This is a small-team, high-ownership Senior Software Engineer role at the intersection of cutting-edge agentic AI and FDA-regulated medical device software. The engineer will build the core AI platform infrastructure that the rest of engineering builds on top of, including LLM-powered tooling to accelerate Heartflow's own software development lifecycle. In-office presence in San Francisco or Santa Rosa is required 3 days per week. Responsibilities - Build and maintain shared libraries, MCP tool integrations, and deployment pipelines powering internal agentic systems - Apply LLMs to accelerate the software development lifecycle — parsing requirements, mapping to modules, and auto-generating test cases - Design Human-in-the-Loop workflows aligned with FDA compliance requirements - Build knowledge retrieval pipelines evolving from RAG toward graph-based and agentic retrieval for clinical and regulatory data - Develop and maintain MCP host environments bridging AI systems with GitHub, Jira, Salesforce, and Slack - Create automated LLM-as-a-judge evaluation harnesses integrated into CI/CD pipelines - Build internal APIs, sandboxes, and developer tooling enabling product engineers to safely integrate AI - Mentor teammates and translate AI concepts for clinical, regulatory, and product stakeholders Requirements - Strong proficiency in Python and TypeScript for end-to-end agentic application development - Deep experience with FastAPI, React (or similar), and AWS with Infrastructure as Code (Terraform) - Production experience building LLM API integrations (Anthropic Claude, OpenAI), prompt engineering, and RAG orchestration - Experience with MCP server/client architecture for deterministic tool integrations - Automated CI/CD-integrated evaluation experience using Langfuse, Braintrust, or Promptfoo - Familiarity with HIPAA and PHI data privacy requirements in healthcare settings - 7+ years of relevant industry experience; Bachelor's degree in CS, Engineering, or equivalent Nice to have - Experience with Go, Rust, or C++ for low-latency API infrastructure - Fine-tuning open-weight models (LoRA, PEFT) or deploying local models with PyTorch or Hugging Face - Hands-on experience with LangGraph, Pydantic AI, or multi-agent frameworks; Playwright browser automation - AI adversarial testing, red teaming, STRIDE threat modeling, or Promptfoo evaluations - Background in MedTech regulatory frameworks (FDA 21 CFR Part 11, IEC 62304, GxP, ALCOA+) Benefits & perks - High ownership and direct influence on an AI transformation at a publicly traded medical device company - Work alongside world-class deep learning researchers and clinical scientists Compensation - Salary not specified. Equity and bonus details not specified.