Software Engineer, Applied AI/Product

We are a small team that cares about deep thinking, speed, craftsmanship, and putting customers first. Our mission is to build the best care experience in healthcare, on a 30 year horizon.

We help physicians achieve independence by making it easier to start, operate, and grow their practices. We believe the future of healthcare will be driven by individual physicians, not by large bureaucratic systems. Lowering the barrier to practice ownership lets doctors deliver better care for everyone.

The founding team brings deep expertise across medicine, product, operations, and engineering, including more than 30 years of experience running independent practices across urgent care centers, surgical centers, and beyond. We have built 0-to-1 products at Meta, Dropbox, Instagram, Coda, and Nomad, and have founded companies with valuations exceeding $1B and more than 100M MAU.

We have strong product market fit. $15M+ ARR in under 15 months.

We are profitable and funded by top tier healthcare investors.

We are about 30 people today and rapidly growing across the company.

You will fit in here if you want to own a product, a process, or a function and take it from 1 to 100.

We’re building an agentic platform at the intersection of medicine and law, transforming complex, high-stakes casework into structured, actionable context. Our system handles patient records, legal workflows, and sensitive decisions, so reliability, clarity, and thoughtful engineering matter deeply.

As a software engineer, you'll work across the full stack: agent architectures, evals, the data and context that feed them, and the surfaces to interact with the platform and its harness.

building and shipping a new workflow that takes over a step of the case process, driven by something a physician or our ops team flagged

designing the eval framework that decides whether an agent's output is reliable enough to ship, while piping the feedback into our synthesized knowledge base

improving how the platform ingests, indexes, and assembles the right case context for an agent to work from

adding new tooling to the interfaces users interact with day to day, bringing more of the platform's context and agent capabilities into their workflow

iterating quickly on a feature with direct input from users, refining the agents based on real-world usage and feedback

setting the patterns and tooling for AI-native development that the rest of the team builds on

engineer for correctness and reliability, not just functionality

are excited to work with frontier models, and care as much about the evals and guardrails around them as the models themselves.

ship high-quality code fast, and can take a vague problem to a shipped solution without a spec.

own problems end-to-end, from the agent logic to the interface a user actually uses.

care about craft and the gap between something that works and something genuinely good to use.

want to play long-term games with long-term people. We're building for a 30-year horizon, not a quick exit.