Senior Full Stack Engineer

Senior Full Stack Engineer

Era Supports · Engineering

Full-time · Individual contributor · *[Remote — TBC]*

*Own the platform that decides which patient in recovery gets a call today, helps shape what is said to them, and makes sure the clinic gets paid for making it.*

About Era Supports

Era Supports exists to change how people recover from substance use disorder. We deliver evidence-based collaborative care inside community health centers, embedding behavioral health care managers alongside primary care teams so that physical, emotional and social needs are treated together rather than in separate silos.

The software is not a side project — it *is* the operating model. Our platform runs the clinical registry, the daily patient outreach engine, assessment scoring and acuity tiering, the billing pipeline for collaborative care, and the integrations with our partners' electronic health records. When a care manager knows who to reach today, or a health center gets reimbursed correctly for collaborative care minutes, that is our code doing its job.

The role

You would be our senior engineer, and you would be building.

This is deliberately an individual contributor role.

The platform is currently maintained by a very small team, and what we need first is someone who ships — who can own significant parts of the system end to end, from schema design through to the deploy and whatever happens afterwards. As the team grows you would have the first claim on shaping how we hire and how we work, if that is something you want. But that is later. Today the job is to build.

The surface area is wide and we are not going to pretend otherwise: a NestJS backend with a domain-driven core, three React applications, a Python scheduling service, an SFTP-based ETL pipeline, a GraphQL integration with an external EHR, and a growing body of AI-assisted clinical tooling. You will not be handed a narrow slice of this.

What you would actually work on

Not hypotheticals — these are live on the roadmap right now.

• AI inside the clinical workflow. Help develop responsible AI used in a clinical setting using evaluation, guardrails, human review, and real limits on what data may inform a model. These are some of the most consequential engineering projects we have.
• Multi-region architecture. Improving our systems and care platform to have full support for partners and billing in multiple regions.
• Deeper EHR interoperability. We integrate with EHRs via API today. Partners increasingly want ADT feeds, FHIR endpoints and health information exchange connectivity. This is protocol work with immediate clinical consequences.
• The outreach engine. A Python service scores every enrolled patient daily against a set of clinical rules to decide who the care team should contact and in what order. Extending and tuning that logic is one of the most direct levers we have on patient outcomes.
• Billing correctness. Collaborative care, behavioral health integration and chronic care management billing all depend on cumulative time tracking, documentation state and payer-specific rules. Mistakes here are expensive and slow to surface.
• Hardening what we have. We are a small team that has moved quickly, and it shows in places — test coverage, authorization edge cases, observability. You would have a mandate to fix that, not merely permission.

Our stack

Backend

TypeScript and NestJS, domain-driven design with CQRS and an event bus, PostgreSQL, Redis and BullMQ for queues and scheduled work

Frontend

React 18, TypeScript, MUI, Zustand, SCSS

Services

Python 3.12 and FastAPI, running on AWS Lambda

AI / ML

LLM-backed clinical tooling — patient messaging, risk stratification and population health focused

Integrations

Multiple EHRs via API and HL7 standard interfaces, Auth0 for identity, SFTP-based ETL for partner data, Metabase for analytics

Infrastructure

Docker, AWS, Aptible, GitHub Actions, Sentry

You do not need to have used all of this. You do need to be genuinely fluent in TypeScript on both sides of the wire, and untroubled by picking up the rest.

What we are looking for

• Substantial production experience building and running web applications, with real depth in TypeScript, Node and React — not a passing familiarity with each.
• Genuine end-to-end ownership. You have designed the schema, written the API, built the interface, and then debugged the whole thing in production when it mattered.
• Judgement about pace. Knowing when to build carefully and when to build fast, and the honesty to say out loud which one you just did.
• Early-stage temperament. You have worked somewhere without a platform team, a QA team or a dedicated SRE, and you treated that as the conditions rather than a grievance.
• You have shipped AI features, not just tried them. You have taken LLM-backed functionality to production and dealt with what that actually involves: evaluation, failure modes, guardrails, cost, latency, and a clear view of where a human has to stay in the loop. This matters to us considerably more than model-training experience.
• Care about correctness where it counts. Some of our bugs are cosmetic. Some of them mean a patient in recovery does not get a phone call, or that a generated message reaches someone it should never have reached. Knowing the difference is most of the job.

Valuable, but not required

• Healthcare experience — particularly behavioral health, collaborative care, or revenue cycle
• Deeper ML or data science background: predictive modelling, feature engineering, or working with claims and clinical data at population scale
• Exposure to value-based care, risk contracts, or population health analytics
• Healthcare data standards: FHIR, HL7v2, ADT, X12
• Working knowledge of HIPAA and 42 CFR Part 2 in practice rather than in principle
• Python, for the scheduling service

What this role is not

• Not a management role, for now. This role is not responsible for managing a team, there may be opportunities to take on team management as we grow.
• Not greenfield. You would be working in a real codebase with real history, some of it considered and some of it the product of a startup moving fast.
• Not ticket-taking. You will play a role in defining the technical implementation and system architecture based upon product specifications in requirements docs and tickets.

How we work

Small team, direct communication, and a short path from problem to production. We work with clinicians every day, which means feedback arrives fast and specific — you will hear about it when a workflow does not fit the way care actually happens.

We take privacy and security seriously because we hold some of the most sensitive health data there is. Substance use records carry protections beyond HIPAA under 42 CFR Part 2, and that genuinely shapes how we design systems rather than being a compliance box at the end.

It also makes applying AI here harder, and a good deal more interesting, than in most domains. What data may inform a model, what a model may be permitted to draft to a patient, and who signs off before it is sent are design inputs from the first line of code — not questions to answer after the demo works.

What we offer

• Meaningful equity and competitive cash compensation
• Ownership of the technical direction of a platform that measurably affects whether people stay in recovery
• Direct access to leadership and to the clinicians using what you build
• Health, dental and vision benefits; paid time off

Pay: $101,000.00 - $123,000.00 per year

Work Location: Remote