Application Developer (Java, JavaScript) (Full-Stack)

Broadway Ventures

5h•$65 - $70

About The Position

The Application Developer builds Java services for Project Argos that ingest public health IT data, monitor deployed FHIR endpoints, validate findings, and assemble reproducible evidence cases. The role develops deterministic validators for pass/fail decisions and implements retry, rate-limiting, and instrumentation logic to make monitoring safe and repeatable. Agentic components may summarize and draft, but they do not make final compliance decisions; the developer’s code enforces that boundary. Project Argos is an agentic AI proof of concept for the Office of the National Coordinator for Health IT (ONC). Its pipeline retrieves published API endpoint data for roughly 290 certified health IT products from the Certified Health IT Product List (CHPL), tests endpoints on a repeatable schedule, and builds reproducible evidence cases for compliance gaps under 45 CFR Part 170. It runs in a contractor-hosted AWS GovCloud environment, and a qualified human reviews every finding before it becomes a confirmed issue. This is a one-year contract, part-time role with variable hours of 10–29 hours per week. The position is open to W2 and 1099 arrangements. Pay is $55–$60 per hour for W2 or $65–$70 per hour for 1099. The developer works with the Technical Lead, AI/Engineering Lead, Senior Systems Engineer, and Project Manager. Applicants must live and work in an eligible U.S. state; the posting excludes California, Colorado, Connecticut, the District of Columbia, Hawaii, Illinois, Maryland, Massachusetts, New Jersey, New York, North Dakota, Ohio, Oregon, Rhode Island, Washington State, and Wyoming.

Requirements

At least 5 years of Java 17+ development experience, including concurrency and networked services.
At least 3 years of experience with Spring Boot or an equivalent framework.
At least 2 years of JavaScript/Node experience for tooling and front-end development.
At least 4 years of experience building REST APIs and working with JSON.
At least 3 years of working knowledge of HTTP and TLS fundamentals.
At least 3 years of experience with the AWS SDK, DynamoDB data modeling, and S3.
At least 2 years of experience building and running containerized services.
At least 4 years of automated testing experience, including JUnit and integration tests.
At least 3 years of code review and technical documentation experience.
At least 1 year of experience building with LLM APIs or using AI coding assistants in development, including testing and validating AI-generated output.
Authorized to work in the United States.
Must live and work in an eligible U.S. state; applicants living in or working from the states and locations listed in the posting are not eligible.

Nice To Haves

HAPI FHIR or other FHIR R4 validation experience, including US Core and CapabilityStatements.
Experience with HTTP client internals, including timing and connection instrumentation.
Experience designing scheduling, retry and backoff, and rate limiting.
Experience with Testcontainers or stub and mock services for failure testing.
SMART on FHIR, Inferno, or ONC certification testing experience.
Health IT or federal program experience.

Responsibilities

Build ingestion connectors for CHPL, Service Base URL Lists, developer-published FHIR Bundles, Lantern, and other approved public sources, preserving raw artifacts, timestamps, request metadata, response headers, and hashes.
Normalize developer, product, version, criterion, CHPL ID, endpoint, organization, and validation records into the project’s common data model in DynamoDB, including lineage and source-of-truth precedence rules.
Implement publication checks that fetch endpoint lists and classify availability failures such as missing URLs, timeouts, obscuring redirects, malformed responses, and non-machine-readable content.
Build safe, non-authenticated endpoint observation for metadata and CapabilityStatement retrieval, and detect endpoint additions, removals, moves, or changes.
Run FHIR R4 validation against Endpoint, Organization, and Bundle resources, preserving raw and normalized results.
Build evidence case packages containing source metadata, request and response logs, validation outputs, severity, and rule traceability; support deduplication, clustering, and export to Government-designated systems.
Design scheduling, retries and backoff, rate limiting, error handling, and audit logging for pipeline and agentic components.
Make test runs replayable so reviewers can rerun tests for any evidence case.
Write unit and integration tests, including stub and mock endpoints for failure scenarios; participate in code review and keep technical documentation current.
Ensure code does not request, collect, or store patient data, PHI, credentials, or tokens unless the Government authorizes it in writing.