Senior Staff, AI Data Engineer

About Stellar Health:

Historically, US Healthcare has relied on a fee-for-service reimbursement system where providers are paid based on the quantity of patient visits and procedures, rather than the quality of health outcomes.

At Stellar Health, we help primary care providers put patient health first. Our platform - a mix of technology, people, and analytics - supports providers at the point of care, delivering real-time patient information, activating practice staff, and empowering providers and care teams with incentives that reward the work they are already doing to keep patients healthy. Using the Stellar App, our web-based, point-of-care tool; practices receive a simple checklist of recommended actions that support the best quality care. Providers and care teams are then paid monthly for each action they complete, and Payors save money in reduced healthcare costs along the way.

Stellar is a US-based Health-tech backed by Top VCs (General Atlantic, Point72, & Primary Venture Partners) with an established product & proven operating model. We’ve shown that we make a real difference for physician practices and their patients.

Senior Staff AI Data Engineer

Stellar Health is hiring a Senior Staff AI Data Engineer to lead how AI transforms analytics and reporting, and to evolve the data architecture that makes it work.

You’ll own, drive, and contribute to two key components of our analytics strategy.The first is AI for analytics and reporting, end to end: strategy, tool selection, administration and architecture, semantic and context layers, and governance. The second is our data architecture. You’ll lead the thinking on how it and our models need to evolve to support these use cases, and contribute to the dbt + Snowflake work to get there alongside our analytics engineers.

Together they serve one goal: compress time-to-insight, and improve the quality of the insights themselves.

What you’ll own

AI for analytics and reporting

• Set the strategy for how AI reshapes analytics, reporting, and self-serve data access at Stellar: where it goes first, what it replaces, what it unlocks, and how we sequence it.
• Own tool selection, administration, and architecture, including Claude on Snowflake, MCP servers, skills, agent workflows, and IDE integrations, along with the reliability and scaling limits as usage grows.
• Design the eval and guardrail systems that catch regressions before they ship and give the team confidence in agent output for high-stakes reporting.
• Define how analytics use cases get proposed, governed, and shipped, so AI access to our data is both impactful and controlled.
• Debug the hairiest failures, including the agent problems that look like LLM problems but are actually data modeling, metadata, or system-design problems.
• Build for futures we can’t fully predict: new agent patterns, new tools, new business questions, new regulatory constraints.
• Be the role model and mentor for AI-assisted analytics across analytics engineers, analysts, and other engineers. Identify what’s worth sharing externally and lead the content and communication around it.

Data Modeling

• Work alongside the analytics engineering team to extend and refine the dbt + Snowflake models that serve reporting, analytics, and AI consumption, bringing particular attention to the structure, grain, and documentation agents require for accuracy, not only what a human analyst can navigate.
• Inform standards for how models, metrics, tests, documentation, and lineage are structured, and build the consensus that gets them adopted team-wide.
• Lead the semantic and context layers: metric definitions, business context, and the metadata that makes both dashboards and agents trustworthy, developed with the analysts and engineers who rely on them.

Target experience

• 8+ years across data platform, ML platform, or developer-tooling - including time as the senior technical voice for a major data/ML/AI surface.
• Production experience architecting LLM systems: agents, evals, RAG, MCP, tool use, multi-step workflows, and the failure modes of each.
• Strong dbt + cloud warehouse (Snowflake preferred) chops; opinionated about semantic layers and metadata.
• Track record of driving wide-team consensus on technical direction - not just shipping, but getting other teams to adopt the pattern.
• Comfortable communicating externally: blog posts, talks, OSS contributions, customer-facing thought leadership.
• Bonus: Claude Co…