Four Roles, Two Lists, One Priority
Medplum's careers page lists five open positions. Its Y Combinator board lists four different ones. The discrepancy reveals how a twenty-person team building an open-source EHR platform prioritizes its next hires.
The YC listings show four roles with defined compensation bands and experience floors: Founding Account Executive, Forward Deployed Engineer, and two engineering roles. The careers page shows Head of Security, Security Engineer, Full Stack Engineer, Forward Deployed Engineer, and Account Executive. Both surfaces agree on Forward Deployed Engineer. Both signal a sales motion. Beyond that, they diverge. A team founded in 2022 — Y Combinator S22 batch, veterans from Microsoft, One Medical, Palantir, and Meta — doesn't post ghost roles. It posts what it needs next.
| Role | Base Salary | Equity | Experience Floor |
|---|---|---|---|
| Founding Account Executive | $200K–$250K (Y Combinator's jobs board reported) | Not listed | 6+ years |
| Forward Deployed Engineer | $120K–$200K (according to Y Combinator) | 0.01%–0.20% | 3+ years |
| Developer Experience Engineer | $150K–$190K | 0.01%–0.20% | 3+ years |
| Software Engineer | $160K–$220K (Y Combinator's data shows) | 0.01%–0.20% | 3+ years |
The Founding Account Executive role carries the highest base and the steepest experience requirement: six years minimum. The equity column is blank on the YC board for this role, but at a twenty-person company post-S22, the grant will be meaningful. This person owns the commercial motion for an API-first EHR that targets digital health builders and life sciences companies. Medplum describes its platform as "trusted by leading digital health and life sciences companies."
Forward Deployed Engineer sits at the widest salary band — an $80K spread — and the lowest base. The role exists because Medplum's product doesn't ship as a finished application; customers build on it. The YC description says FDEs "conduct enterprise workshops and trainings, troubleshoot customer implementations, provide technical guidance and create documentation" and "write code for core healthcare workflows, such as e-prescribe, labs, health networks, billing, benefits check, etc." Three years experience is the floor. The equity range (0.01%–0.20%) matches the other engineering roles.
Developer Experience Engineer and Software Engineer share the same equity band and experience floor. Their bases differ by $10K at the bottom and $30K at the top. The DXE role emphasizes "SDKs, docs, CLI tooling" and the "time to first FHIR resource" metric per Medplum's roadmap. The Software Engineer role owns the platform core: the open-source EHR engine, compliance boundaries, and multi-tenant architecture.
The careers page tells a different story. Head of Security and Security Engineer appear there, not on the YC board. Full Stack Engineer replaces the Software Engineer/DXE split. Account Executive replaces Founding Account Executive. The company hasn't clarified which source is authoritative. Candidates should check both.
A fifth role, Business Operations Associate, appears on TheJobsMap and a dedicated Medplum careers page at $90K–$125K base, entry-level. The description reads like a chief of staff track: cross-functional orchestration, compliance program support (SOC II, HITRUST), partner management for labs, RCM, medication, and fax vendors, event logistics, sales enablement, and "Canon" knowledge management (Medplum's term for "standardized written docs, slides, and code that serve the entire organization.") The role demands extreme ownership, analytical rigor, and slide-craft.
Four roles on the YC board. Five on the careers page. One on a third-party aggregator. The count depends on where you look. What doesn't change: every opening sits at the intersection of open-source infrastructure, healthcare regulation, and developer experience.
Inside the Funnel: No Portal, No Bots
Medplum's hiring funnel starts with a single, deliberate gate: an email to [email protected] with a resume and a short introduction. No applicant tracking portal, no automated form, no "easy apply" button. The careers page and the "We're Hiring" blog post both route every candidate through that address.
From there, the process is not publicly documented in granular stages. The company repeats three themes across its channels: engineering excellence, open-source fluency, and mission alignment. The "We're Hiring" post states the team is "committed to maintaining a tight team of elite engineers" and that they "use modern tools and follow industry best practices." The careers page adds that they "take pride in our engineering work, documentation and aspire to engineering excellence and best practices" and value "continuous learning and improvement and working in public."
The team's pedigree sharpens that rubric. Founders and early members come from organizations where code review, design docs, and operational rigor are baseline expectations: Microsoft, One Medical, Palantir, Y Combinator, and Meta.
The VP of Engineering role makes the internal standard explicit: "Ensure technical excellence and the adoption of best practices as we grow," "Help establish and formalize engineering best practices," and "Be the engineering representative for planning, triage, and customer engagements."
For the security roles (both positions), the filter adds a compliance dimension. Medplum's platform is an open-source, API-first EHR. The marketing language is precise: "skip the plumbing and ship what matters, getting you to market faster with a secure, compliant, and scalable foundation." The Account Executive role, the lone non-engineering opening on the careers page, will be screened for enterprise health-tech sales cycles.
What happens after the initial email? The research does not publish a stage-by-stage flowchart, and Medplum has not released a public interview guide. But the company's "building in public" ethos extends to its hiring signal: they invite outreach even when no role matches, and they emphasize the short introduction. That introduction is the first work sample.
Medplum does not use LeetCode-style puzzles. Industry research ("Technical Assessment Best Practices 2026") argues that LeetCode doesn't predict job success and that top companies have moved to pair programming, take-home projects, and AI-aware evaluation. Medplum's "modern tools and industry best practices" line, combined with its open-source DNA, aligns with that shift.
The filter is tight by design. "Tight team of elite engineers" means each hire raises the bar for the next.
What the Stack Demands
Medplum's architecture reads like a checklist of the hardest problems in healthcare interoperability. The platform's seven core components (FHIR R4 Data Store, RESTful API + GraphQL, Subscription Engine, Bot Framework, Identity and Access Management, React UI Component Library, Integration Layer) each demand a different flavor of expertise.
FHIR R4 isn't a nice-to-have at Medplum; it's the storage format. The company stores clinical data in native FHIR R4 rather than converting on demand. As Medplum's documentation notes, "FHIR-native development reduces avoidable translation. It does not eliminate healthcare data modeling." The platform's ONC certification (CHPL #11745, 19 criteria including (g)(9) and (g)(10)) and Cures Act compliance mean regulatory literacy carries weight equal to algorithmic skill.
On the engineering side, the stack is unapologetically TypeScript and Node.js. The Bot Framework runs server-side logic in TypeScript; the React component library expects contributors comfortable with modern hooks, state management, and accessibility patterns. Self-hosting documentation explicitly calls out "strong cloud, Node.js, and command-line proficiency" as prerequisites for production deployments. Half a million Docker downloads suggest container fluency is assumed. GraphQL appears alongside REST in the API layer.
The Integration Layer and Medplum Agent reveal a third skill set: legacy protocol fluency. HL7 v2 message processing (ADT, ORU, ORM) over MLLP, ASTM for lab instruments, DICOM for imaging, and SFTP for batch exchange are the daily plumbing that connects Medplum-hosted applications to hospital IT. The Agent runs on-premises to bridge bedside monitors and non-FHIR equipment. FHIRcast enables real-time context synchronization between applications.
Open-source contribution history carries signal because Medplum is Apache 2.0 licensed with over two thousand GitHub stars. The 2026 roadmap includes a plugin ecosystem ("extend Medplum without forking").
Compliance competence rounds out the profile. HIPAA BAA management, AES-256 encryption at rest, TLS 1.3 in transit, SOC 2 Type II readiness, and CLIA-ready lab workflows are table stakes. HITRUST certification is on the 2026 roadmap. The noted gap (Medplum certifies on US Core STU 5.0.1 while USCDI v3 mandates US Core 6.1.0) tells you what the team is hiring for: engineers who can close that version delta without breaking existing customer implementations.
Reading the Roadmap in the Reqs
The openings Medplum posted this cycle map to declared 2026 roadmap pillars. The cluster tells a clearer story than any press release: the company is moving from "open-source infrastructure" to "production-grade platform" in a single fiscal year.
Start with the backend hires. Medplum's 2026 roadmap lists three infrastructure bets that require deep systems work: sharding and archiving for large datasets, Kafka-backed data-warehouse integration, and cross-cloud support across GCP, Azure, and AWS. The 2025 numbers already show strain: 500,000+ Docker downloads (double the prior year), 161 GitHub contributors (up 40%), and a hosted-project count that tripled to 3,700. The maintainer team sits at 11.
The Developer Experience Engineer role signals a shift in go-to-market. Through 2024 Medplum grew its Discord to 1,076 users (13×) and GitHub stars to 2,100 largely through organic pull requests and FAQ threads on Linen: questions about token refresh across tabs, pagination limits on large datasets, migration scripts from v3 to v4. The 2025 roadmap explicitly calls for a Technical Advisory Board and the first PlumCon in San Francisco.
The Forward Deployed Engineer hire is the enterprise signal. Medplum's provider experience is already ONC-certified as a base EHR. The 2026 roadmap adds HITRUST certification for the hosted environment, HTI-4/CMS-0057-F compliance (January 2027 enforcement), integrated medical billing, prior-authorization automation, and a plugin ecosystem for revenue-cycle tooling. Wire compatibility with Epic, Cerner, and Athenahealth means data generated on Medplum exchanges with those systems without custom translation layers. The FAQ history shows real deployment friction: customers asking about Firebase ID token exchange, multi-tab token refresh bugs, scheduling data-model recommendations.
| Roadmap Pillar | 2026 Milestone | Role That Owns It |
|---|---|---|
| Scale & Performance | Sharding, archiving, Kafka, cross-cloud (GCP/Azure/AWS) | Software Engineer / Full Stack Engineer |
| Regulatory Readiness | HITRUST, HTI-4 compliance by Jan 2027 | Forward Deployed Engineer |
| Developer Experience | Scheduling, billing, AI tooling, plugin ecosystem | Developer Experience Engineer + Backend |
| AI Infrastructure | MCP, SSE, websockets, composable model configs | Software Engineer |
| Provider Workflows | Self-scheduling, resource scheduling, encounter docs, RCM | Forward Deployed Engineer |
Watch the GitHub project board for confirmation. Current WIP tickets — Node.js 26 in CI, TypeScript 7 upgrade testing, Vitest parallelization, OpenTelemetry instrumentation fixes, hypopg hypothetical-index analysis for EXPLAIN plans — are exactly the infra hygiene a team preparing for sharding and multi-cloud would prioritize.
How to Get Past the Screen
Medplum's job descriptions read like a filter, not a wish list. The company is 20 people (veterans from the aforementioned companies), and every open role signals a specific gap.
** The stack is Node.js, React, TypeScript, but the product is an API-first EHR built on FHIR.
It hires from the roadmap outward, and the roadmap says scale, compliance, and developer experience are the three problems worth solving next.
**Lead with the mission, not the stack. ** The careers page repeats it twice: "increase access to and utility of medical data" and "impact the lives of patients, speed medical research, and contribute to the open source ecosystem." In a cover letter or intro call, connect your past work to that mission concretely. The Forward Deployed Engineer role explicitly wants someone who can "build relationships with customer executives and become a trusted thought partner for the C-Suite."
**Show, don't tell, on "working in public." ** The team values "engineering excellence and best practices," alongside "working in public." That means open-source contributions, technical writing, conference talks, or thorough GitHub READMEs on side projects count as signal. The Forward Deployed Engineer will "create documentation to help customers unlock value faster" and "be the voice of our customers" to product and engineering.
**For the Forward Deployed Engineer: prove autonomy and commercial instinct. ** The JD uses phrases like "win at all costs mentality," "self-managed," and "sweat the details." This is the founding member of Solutions Engineering, responsible for "reducing time to value," "growing existing customer value," and "developing enterprise features and integrations" like e-prescribe, labs, and billing workflows. The salary band ($120K–$200K base, 0.01%–0.20% equity) reflects a senior IC who operates like a founder.
**Security roles: speak healthcare compliance fluently. ** These security roles aren't generic infra roles. Medplum handles PHI for "some of the largest digital health companies and life sciences companies." The team's Palantir and One Medical alumni know the difference between compliance theater and actual data protection.
**Full Stack Engineer: depth over breadth on the stack. Experience with healthcare data models (resources, bundles, search parameters, compartments) separates candidates who will be productive in week two from those who need month three. If you've contributed to open-source tooling in this space (Medplum's own SDK, HAPI FHIR, Firely), reference it. The team "aspires to engineering excellence."
**No perfect match? Email [email protected] anyway. ** The careers page explicitly invites this: "Even if you don't see a position that perfectly matches your skills, don't hesitate to reach out."
**Prepare for a process that tests collaboration, not just code. ** With a team this small and senior, every hire changes the culture. The "continuous learning and improvement" value means they'll probe how you handle being wrong, how you review code, how you document decisions.
The Signal in the Discrepancy
The YC board and the careers page still disagree on the count. The YC list shows four roles with compensation bands; the careers page shows five with security and a generalist engineer. The third-party aggregator shows a sixth. Medplum hasn't reconciled them. But the disagreement itself is the clearest signal: a team this small, this senior, and this close to the metal doesn't hire from a template.
The candidates who clear the screen won't be the ones with the longest keyword lists. They'll be the ones who can point to a FHIR resource they modeled, an HL7 integration they untangled, or a piece of documentation that outlasted their tenure. The email address hasn't changed. The bar hasn't lowered. The next twenty people will write the next chapter of the platform — in public, with rigor, and with the receipts to prove it.
Working in frontier tech? Zero G Talent tracks the openings: see every open ASML role, browse frontier tech jobs, openings at Stripe, and the people building the field.