About the role
- You will build IRIS, 5C Network's Radiology Operating System for large multi-site hospitals: RIS, PACS integration, reporting and operations on one platform. You will own its production architecture and take it live in hospitals.
- 5C has built teleradiology in India since 2017. Today we read 20,000+ scans a day, with 400+ specialist radiologists working in our ecosystem. All of it runs on software we built ourselves: worklist, routing, AI pre-read, reporting, QC and turnaround control. IRIS turns that system into a product hospitals run themselves. It isn't a guess at what hospitals need; it's built on years of running radiology at scale.
- Our first deployment is with one of India's leading hospital chains, across multiple hospitals and a central hub. A working sandbox (the MVP) is already live.
- Bengaluru, on-site, 6 days a week. 2–4 years of experience.
What you'll build
- Hospital integrations: HL7 orders coming in from the hospital information system and reports going back out, plus DICOM worklist sync with scanners across sites
- Real-time worklists: routing studies between hospitals and the hub, with turnaround timers that doctors and ops teams act on live
- Roles and permissions for 8+ clinical roles: receptionist, technician, nurse, medical officer, residents, consultants and QA, each with its own workflow
- On-prem deployments that keep running when hospital networks don't
- AI in the clinical loop: turning AI pre-reads into something radiologists trust, with evals, feedback loops and audit trails
- Production-grade reliability: observability, audit logs and uptime a hospital can depend on
Your first 90 days
- Week 1: shadow technicians and radiologists in a live radiology department
- Day 15: ship your first feature to the sandbox, in front of real users
- Day 30: own a full module end to end: design, build, deploy and support
- Day 60: ship your module to production in a live hospital
- Day 90: run your module in production across sites, and help shape the IRIS roadmap and technical decisions for the next phase
Must have
- Strong full-stack skills: React and TypeScript on the frontend, Node.js or Python on the backend
- Solid PostgreSQL and API design. Comfortable with queues, caching and containerized deployment (Docker)
- You've shipped production software that real users depend on, and owned it end to end
- AI-native: you use Claude, Cursor or similar daily to build faster, and you know where to trust it and where not to
- High ownership. You can take a rough mockup and a clinical workflow and ship it without waiting
- You've worked on B2B software products end to end, ideally 0-to-1, with real customers
Good to have
- Healthtech experience (RIS, PACS, HIS, EMR) or exposure to DICOM, HL7 or FHIR
- Auth and permissions (Keycloak, policy engines), on-prem deployments, observability
- Experience with workflow-heavy enterprise software used by hospitals or operations teams
Why join
- Building AI into high-stakes clinical workflows, not demos
- Real scale from day one: a business already reading 20,000+ scans a day, a live MVP and a committed first customer. Not a blank page or a pitch deck
- Architecture decisions most engineers wait 8+ years to make
- Work directly with the founder. As IRIS grows, you'll help hire and lead the next engineers, with a clear path to tech lead
- AI-native team: paid Claude and Cursor seats, and the freedom to shape how the team builds with AI
- Product judgment: you'll work directly with radiologists, hospital ops teams and the founder, and join customer demos and go-lives
What this is not
- Not a ticket-taking role on a mature codebase
- Not a frontend-only or backend-only role
- Not a role where you wait for specs
