Senior Developer - Developer Acceleration at Jane App

Job description

About The Role Hi, I'm Devlon . I lead Developer Acceleration at Jane, joining back in February. My team's job is simple to say and hard to do well: give Jane's developers measurable leverage to ship faster. Sometimes that's a tool someone invokes, sometimes it's a foundation they build on without even noticing it's there. Either way, our "customers" are the ~300 developers across Jane, and our work is what lets them spend less time fighting friction and more time building the things that help our clinics run their businesses and care for their patients. That's the mission, just one layer removed: we help the people who help the helpers. This role matters because we are at the point where AI-augmentation for product development should be a paved path. You'd be one of the people building that path, not just for one team, but for the whole engineering org. What Impact We're Looking For You To Make Build the tools and shared infrastructure that make every developer at Jane faster. That spans the agentic context and harness the monolith relies on, the plugins and skills teams reach for daily, the environment that lets agents do real work safely, and the measurement that tells us what's actually landing. Your work multiplies across the whole engineering org, not just one product surface. Bring a systems eye to how developers actually work by embedding with product teams for short, focused stints. Sit in their stand-ups, ship a ticket from their backlog, and find where time actually goes (CI noise, oversized PRs, review latency, changes-requested churn). Leave with a diagnosis and a first fix, not a report. Then turn what you saw into tooling that removes the friction rather than papering over it, and that spreads beyond the team you sat with. Run each initiative like a small startup. Start from a real pain, ship the smallest version that tests whether it matters, and measure whether developers come back to it. Grow it when more than one team pulls for it. Kill it when they don't. Most of what we try should be cheap to drop. Build with a platform mindset, and the judgment to know when it applies. Most of what you make should plug into a larger system we own and maintain, whether that's running agents in the cloud, distributing skills to developers, or managing context. But a quick one-off is sometimes the right way to learn what's real before committing to the durable version. The skill is knowing which is which, and when a fix that worked for one team has earned a place in the shared system. Drive agent-authored code from novelty to normal, growing the share of PRs that agents open, land, and merge. Your own workflow is the starting point, not the goal. The job is turning what works for you into patterns, skills, and defaults other teams adopt without being told to. Communicate early and often in team channels, sharing context, surfacing risk, and helping everyone stay aligned in a distributed environment. We lean toward over-communicating here, and that's on purpose. What Experience We Need Deep, hands-on experience designing systems other developers depend on, with the judgment to know when tooling helps and when it gets in the way. Hands-on, daily experience building with agentic AI coding tools (Claude Code, Cursor, and similar) across the software development lifecycle. This is the center of the role, not a nice-to-have. We're looking for someone who already delegates real work to agents: writing specs, running multi-file changes, and reviewing output critically. A track record of building production-grade systems, often from the ground up, paired with real understanding of what it takes to work across large, long-lived codebases with many teams and deep history. That includes the particular problem of getting an LLM the right context to be useful inside all of it. Genuine curiosity about developer experience and productivity, and the understanding that building a tool is only half the job. The other half is getting it used: t