Software Engineer III - Developer Platform & AI Enablement

GHX is seeking a Software Engineer III to join the Developer Platform & AI Enablement team, which builds the tooling, standards, and AI capabilities that power the core product engineering at GHX. Reporting to the Director of Software Engineering, you’ll collaborate with a small team whose customers are other engineers—those building healthcare supply chain products that save over a billion dollars annually in healthcare delivery costs.

The Software Engineer III is responsible for analyzing, designing, programming, debugging, and modifying software in an Agile/Scrum environment. You’ll grow into owning a capability end-to-end, such as a surface of the GHX agent platform, an MCP server relied upon by other teams, or the complexity-based router that determines whether a request runs on a local model or a hosted one. This is an embedded role where you’ll spend most weeks inside a product engineering team’s repository, proving out a pattern before bringing the reusable version back to broader engineering teams.

Duties and Responsibilities

• Analyzes, designs, programs, debugs, tests, documents, releases, and supports platform capabilities end-to-end.
• Authors technical designs and architectural artifacts, and reviews the designs of peers.
• Develops, versions, and documents reusable GitHub Actions and composite actions used across the organization.
• Develops, versions, and documents reusable Infrastructure as Code (IaC) modules (e.g., Terraform) used across the organization.
• Builds and operates MCP servers over stdio and remote/HTTP transports, owning tool schema design, transport selection, and authentication.
• Extends the internal agent platform, including the SDK and APIs other GHX teams build against, and contributes to local model routing.
• Reads and safely modifies unfamiliar codebases, including Java and Maven builds, without owning the application code.
• Drives defects and blockers to resolution, and writes automated tests and system test specifications.
• Instruments developer-facing tooling to measure adoption and extends the team’s delivery-metrics pipeline.
• Participates in architectural design spikes, project planning, and estimation in an Agile/Scrum environment.
• Co-facilitates recorded AI in Automation Office Hours remotely each week.
• Facilitates live sessions, both remote and in-person, as an essential function of this role.
• Actively uses AI-powered development tools (e.g., Claude Code, Copilot, Cursor) to improve productivity and code quality, and authors the agent configuration artifacts other engineers reuse.
• Creates and maintains documentation as part of development and enablement efforts.
• Stays up-to-date on industry trends and best practices related to Snowflake, Python, and AWS.
• May mentor and onboard new team members as needed.

Required Qualifications

• A BS or MS Degree in Computer Science.
• 3-5 years of software development experience.
• Experience building internal tooling whose users were other engineers, with the ability to explain why a given tool was or was not adopted.
• Daily use of agentic coding tools — Claude Code (preferred), Copilot, Cursor, or similar.
• Direct experience authoring GitHub Actions at the reusable workflow or composite action level, or equivalent CI that other people consumed.
• Direct experience authoring Infrastructure as Code.
• Technical writing that other engineers act on — design documents, ADRs, runbooks, or documentation you can point to.
• Willingness to teach live and unscripted, including co-facilitating recorded remote sessions and full-day in-person workshops.
• Willingness and ability to travel as needed.
• Must be able to work independently and as part of a team on multiple overlapping projects.

Preferred Qualifications

• Has measured a non-deterministic system and argued from the numbers — an eval harness, a benchmark, or a rubric.
• Experience building and operating MCP servers, including remote/HTTP transports and authentication.
• Comfortable deploying and debugging on AWS.
• An agent configuration artifact you authored that another engineer used — a skill, subagent, md/AGENTS.md, or MCP tool.
• Experience with local or open-weight inference: Ollama, llama.cpp, MLX, quantization tradeoffs, context and KV-cache sizing, and OpenAI-compatible shims.
• Experience measuring developer outcomes: adoption instrumentation, DORA-style delivery metrics, and data plumbing behind messy organizational data.
• Experience with the healthcare supply chain, EDI documents and standards, an added advantage.