AI Engineer | Agentic Systems
$130k - $200k • Remote • Remote - US
Posted 5mo ago
Job Location
Remote - US
Tech Stack
Remote Work Policy
Fully remote
Categories
AI Agent Engineer
About the job
Machinify is seeking an L4 AI Engineer to build production-grade agentic systems for auditing medical claims. This role involves designing and implementing end-to-end systems that read medical records, reason over guidelines, and produce defensible findings. You will be responsible for defining the system architecture, choosing agent topologies, and developing robust evaluation strategies. The position requires a deep understanding of applied AI, particularly in the context of noisy, long-context documents, and a strong ability to leverage AI tooling to accelerate development.
Responsibilities
- Design agent systems from first principles, including loop mechanics, tool integration, context strategy, and evaluation harnesses.
- Determine single-agent vs. multi-agent topologies and choose between LLM reasoning and deterministic post-passes.
- Engineer the context by optimizing context windows, tool surfaces, structured outputs, and citation grounding.
- Drive evaluation rigor by building evals before agents, diagnosing failures, and proving fixes improve metrics.
- Utilize AI tooling such as Claude Code and Codex extensively for planning, scaffolding, refactoring, and debugging.
- Develop domain expertise in healthcare claims, coding guidelines, and medical records.
Requirements
- 2-4 years of applied ML/AI engineering experience with a Bachelor's in CS, Math, Engineering, or equivalent, or a Master's with no prior industry experience.
- Experience owning at least one production-quality system end-to-end.
- Strong Python engineering skills, including clean abstractions, type discipline, async programming, and tested code.
- Deep, hands-on understanding of agent loops and how models interact with tools.
- Hands-on experience with at least one major agent SDK (e.g., OpenAI Agents SDK, Anthropic SDK, LangGraph).
- Working knowledge of how modern coding agents are built and manage context.
- Fluency with AI coding assistants like Claude Code/Codex as a power user.
- Solid command of VS Code and Git, including branches, rebases, worktrees, conflict resolution, and PR workflows.
- A bias toward measurement and reproducible results.