[Remote] AI Engineer | Agentic Systems

Machinify is hiring a remote AI Engineer to design and build production-grade agentic systems that audit medical claims. The role focuses on architecting agent loops, engineering context and prompt strategies, and driving evaluation rigor while leveraging AI coding tools like Claude Code and Codex. Ideal candidates have strong Python engineering skills and hands-on experience with agent SDKs such as OpenAI Agents SDK, Anthropic SDK, or LangGraph.

• Design agent systems from first principles: loops, tools, context strategy, and evaluation harness.
• Engineer context and prompts, including structured outputs, citation grounding, and effective tool surfaces.
• Build evals before agents; diagnose failures and prove fixes with metrics.
• Use AI tooling like Claude Code and Codex to plan, scaffold, refactor, and debug work.
• Develop domain expertise in healthcare claims and medical records.

• 2–4 years of applied ML/AI engineering experience (or Master's with relevant coursework) and at least one production-quality system owned end-to-end.
• Strong Python engineering: clean abstractions, type discipline, async, tested code.
• Deep understanding of agent loops and hands-on experience with at least one major agent SDK (OpenAI Agents SDK, Anthropic SDK, LangGraph, or equivalent).
• Working knowledge of modern coding agents and context/prompt engineering.