Staff AI Engineer, Agentic Systems
New
M
MachinifyHealthcare AI
Location: Remote - US; Work from anywhere in the US!Full-TimeStaff
Salary not disclosed
Apply NowOpens the employer's application page
Job Details
- Experience
- 8+ years of software or ML engineering experience, including 3+ years building applied ML / AI systems in production
- Required Skills
- PythonGitGenerative AI
Requirements
- Have 8+ years of software or ML engineering experience, including 3+ years building applied ML/AI systems in production.
- Hold a bachelor's degree in computer science, math, engineering, or a related quantitative field; a master's or PhD is preferred.
- Have set the architecture of a production system used by other engineers and led work across a team.
- Bring hands-on generative AI experience, including training or fine-tuning generative models or building production agentic systems.
- Understand agent loops, including tool calls, how tool results re-enter context, loop termination, and failure modes.
- Have hands-on experience with a major agent SDK, such as OpenAI Agents SDK, Anthropic SDK/claude-agent-sdk, or LangGraph.
- Be fluent with Claude Code or Codex as a power user and understand how coding agents are built and engineer context.
- Have strong Python engineering skills, including clean abstractions, type discipline, async, and tested code.
- Have solid command of VS Code and git, including branches, rebases, worktrees, conflict resolution, and PR workflows.
- Have built reproducible evaluations that a team relies on and use measurement to guide shipping decisions.
- Preferred: experience with long-context, citation-grounded systems; production LLM caching, observability, and cost and latency control; or structured outputs and tool interfaces using Pydantic or JSON Schema.
Responsibilities
- Own the architecture of claim-audit agents, including agent loops, tools, context strategy, and evaluation harnesses.
- Define measurable expert-level accuracy with clinical and coding experts and build the evaluation infrastructure the team uses.
- Diagnose system failures, address root causes, and verify fixes against evaluation metrics.
- Engineer systems for long, noisy medical records, citation-grounded findings, structured outputs, and reliable tool use.
- Turn open-ended problems into technical plans and lead projects spanning several engineers and quarters.
- Make decisions about model choice, build versus buy, and cost, latency, and accuracy trade-offs.
- Own reliability, observability, and cost control for production LLM workloads.
- Mentor engineers, review designs and code, and set standards for agent design, evaluations, and engineering quality.
View Full Description & ApplyYou'll be redirected to the employer's site
