Rad AI builds generative AI software for radiology that automates impression generation, integrates with voice-recognition and EMR workflows, reduces dictation workload, and automates incidental-finding tracking and follow-up.
ABOUT RAD AI At Rad AI, we’re on a mission to transform healthcare with artificial intelligence. Founded by a radiologist, our AI-driven solutions are revolutionizing radiology—saving time, reducing burnout, and improving patient care. With one of the largest proprietary radiology report datasets in the world, our AI has helped uncover hundreds of new cancer diagnoses and reduced error rates in tens of millions of radiology reports by nearly 50%. Rad AI has secured over $140M in funding, including a recently oversubscribed Series C ($68M round) led by Transformation Capital, bringing our valuation to $528M. Our investors include Khosla Ventures, World Innovation Lab, Gradient Ventures, Cone Health Ventures, and others—all backing our mission to empower physicians with cutting-edge AI. Our latest advancements in generative AI are used by thousands of radiologists daily, supporting more than one-third of radiology groups and healthcare systems and nearly 50% of all medical imaging in the U.S. at partners including Cone Health, Jefferson Einstein Health, Geisinger, Guthrie Healthcare System, and Henry Ford Health. Recognized as one of the most promising healthcare AI companies by CB Insights and AuntMinnie https://www.radai.com/news/auntminnie-recognizes-rad-ai-omni-reporting-as-2023s-best-new-radiology-software, and ranked by Deloitte https://www2.deloitte.com/us/en/pages/technology-media-and-telecommunications/articles/fast500-winners.html as the 19th fastest-growing company in North America, we are building AI-powered solutions that make a real impact. Most recently, Rad AI was named to CNBC’s Disruptor 50 https://www.cnbc.com/2025/06/10/2025-cnbc-disruptor-50-see-the-full-list-of-companies.html list, highlighting the innovation and momentum behind our mission. If you’re ready to shape the future of healthcare, we’d love to have you on our team! Radiologists don't work in a single application — they move between a PACS, a worklist, and our reporting platform, and they expect all of them to stay in lockstep. This role owns those integrations. You'll build and standardize the real-time context synchronization and data exchange that make Rad AI's reporting platform feel native inside any radiology environment, whether that's a modern FHIRcast session or a decades-old system that only speaks XML dropped in a folder. As a Staff Software Engineer for Reporting Integrations, you'll set the technical direction for the platform those integrations are built on — the foundation that lets us support new partners and new types of systems without rebuilding from scratch each time. Integrations are where our software meets each customer's existing systems, and getting them right often decides whether a customer goes live in weeks or months. WHAT YOU'LL BE DOING - Design, build, and own real-time context synchronization integrations — primarily FHIRcast for context sharing, alongside file-based (XML/JSON drop) integrations with PACS, worklists, and EHRs. - Help architect a flexible, evolvable integration platform — one where new integrations are composed from reusable building blocks rather than rebuilt from scratch, so we can adapt quickly to the needs of a growing and diverse customer base. - Work hand-in-hand with Product to shape what's possible with third-party systems — bringing an integrations-first engineering perspective early, when we're weighing the challenges and opportunities of building something new. - Extend that platform into new categories of integration partners beyond the traditional reporting workflow — non-clinical analytics tools, CDI (clinical document integrity) tools, and clinical AI tools that generate draft reports, among others. You'll design the frameworks that make each new class of integration repeatable. - Collaborate across a wide set of stakeholders including internal teams, customers, external vendors and integration partners to align on how these integrations should work and where they should head. - Partner directly with customers and vendors: get on calls, map out their clinical workflows and edge cases, and carry integrations through go-lives and escalations into production. - Coordinate the connectivity that gets cloud-delivered integrations live by working with hospital IT, security, and identity teams on firewall access, certificate handling, and SSO. - Drive standardization: turn one-off, customer-specific integrations into reusable connectors that shorten every future go-live. - Own reliability and observability for integrations including build the logging, tracing, and staging/E2E testing that make issues diagnosable without re-engaging customers and vendors to reproduce them. - Debug and resolve integration issues end-to-end, often across systems you don't control, using logs, traces, and a strong mental model of how the pieces fit together. - Raise the bar for the team: document hard-won knowledge about vendor quirks and workflow scenarios, and mentor other engineers working in this space. WHO WE'RE LOOKING FOR - A backend leaning full stack engineer who is passionate about integrations in healthcare. This is a niche, and we want someone who wants to go deep in it. - Expertise in part of this stack (TypeScript, Node, React, Electron), with interest in picking up the gaps. - Experience building and operating production services in a cloud environment. - Practical hands-on experience with healthcare context-sharing and data-exchange technologies like FHIRcast, SMART on FHIR, FHIR, or equivalent standards and protocols. - A track record of shipping and supporting production integrations with third-party clinical systems. - Experience thinking at the platform level by building integrations as reusable, extensible frameworks rather than a pile of one-offs. - Strong communication skills where you can run a technical call with a customer, vendor, or partner, diagnose the problem, and propose a standardized path