Staff Software Engineer, Applied AI Team
Summary
Staff full-stack engineer owning the application layer of Fabric's live patient-facing clinical product — an embeddable React widget, clinician portal, and Python service layer — that supports its flagship AI Intake system. You'll shore up foundations, drive architecture, build internal tooling, ship versioned shared components, and partner with AI/ML engineers.
About the Role
What You'll Do
- Month 1: Dive into the widget, portal, and service layer to become an owner of all of them, shipping meaningful changes and forming a tight working relationship with our other engineers and partners.
- Month 3: Take full ownership of a major integration or platform initiative and step into the role of primary reviewer for all application-side design decisions.
- Month 6: Deliver an end-to-end system that external teams depend on, and mentor potential junior engineers to the point where they independently own technical surfaces of their own.
- Build the AI Plumbing: Operate the critical application work that supports AI Intake, including encounter data flow, integration against partner platforms, and every surface a new customer touches (typically managing one in flight and another right behind it).
- Own Internal Tooling: Manage and scale the growing set of internal tools that your own team and our clinicians depend on.
- Drive Full-Stack Architecture: Advise on and make foundational architectural decisions, going deep enough into the system to independently identify what needs shoring up and where development should head next.
- Partner with AI Engineers: Keep the application layer tightly coordinated with the AI/ML engineers who own the conversation pipeline, prompts, and clinical-safety layer.
- Protect Shared Surfaces: Ship and version components that other teams consume as a dependency, explicitly protecting those consumers from expensive breaking changes.
- Maintain Operational Excellence: Keep dependencies current and operate what you build so maintenance stays a daily habit rather than a quarterly crisis.
Why You Might Be a Good Fit
- You are drawn to systems you can own and live with, taking real satisfaction in making quickly built foundations durable rather than just shipping and handing them off.
- You care deeply about the patient on the other end, understanding the real safety stakes of a live clinical product and wanting your work to matter.
- You are a fluent full-stack engineer who moves easily from a React widget down through a Python service and its data layer, making sound architectural calls across the entire stack.
- You are comfortable with agentic coding tools and spec-driven development practices.
- You act as a genuine partner to data scientists, knowing enough about LLM application development to speak the language of prompts, evaluation, and latency tradeoffs without needing to own the models yourself.
- You think in dependencies and blast radius, understanding why breaking changes are expensive and designing shared surfaces with consuming teams in mind.
- You write clearly, mentor generously, and treat making other people's work easier as a core part of the job.
This Might Not Be The Right Fit If...
- You want to design the models and prompts yourself (this role strictly partners with the people who do that).
- You prefer greenfield builds that you hand off, rather than inheriting and shoring up systems other people started quickly.
- You prefer to specialize narrowly in the frontend or backend rather than owning the full stack architecture.
- You view keeping dependencies current, operating what you ship, and mentoring junior engineers as overhead.
Your Qualifications
- 6+ years of professional full-stack web software experience, including at least 1 year operating at a Staff level and living with architecture you built.
- Depth in TypeScript and React, with fluency in a modern server-rendered framework (we use Next.js).
- Strong Python experience with an async web framework, plus the surrounding data layer (relational schema design, migrations, caching).
- Experience building versioned dependencies consumed by other teams (widgets, SDKs, or component libraries).
- Fluency with LLM application development to effectively partner with data scientists on prompts, evaluations, and cost tradeoffs.
Bonus Points
- Healthcare experience involving PHI, HIPAA, clinical workflows, or patient-facing surfaces with real safety stakes.
- Familiarity with agent or workflow orchestration frameworks like LangGraph.
- Experience designing for accessibility and internationalization from the start.
- AWS background (especially Bedrock) and infrastructure-as-code.
- Eval tooling experience (Braintrust, LangSmith, or custom-built equivalents).
- Having been the first or second engineer on a greenfield product and having to live with it.