Staff AI Engineer

📋 Description

• Create agent systems starting from foundational principles. Determine the loop, tools, context strategy, and evaluation framework. Make choices between single-agent and multi-agent architectures, LLM reasoning versus deterministic post-passes, and retrieval versus direct context loading — all supported by data.

• Engineer the context meticulously. The most challenging aspect of developing a successful agent lies in the prompt's input and output. You will focus intensely on context windows, tool surfaces, structured outputs, citation grounding, and the prompt itself.

• Promote rigorous evaluation. Construct evaluations prior to agent development. Identify failure points, address root causes, and demonstrate that the fixes improve the metrics.

• Utilize AI tools like a seasoned professional. A significant portion of your day will involve using Claude Code, Codex, and similar tools to plan, scaffold, refactor, and debug your projects. We anticipate you will be more efficient with these tools than most engineers are without them.

• Become a domain authority. Understanding healthcare claims, coding guidelines, and the medical record is essential. Strong engineers who immerse themselves in the domain become significant contributors in this space.

⛳️ Requirements

• 2–4 years of hands-on ML / AI engineering experience, holding a Bachelor's degree in CS, Math, Engineering, or a related field — or a Master's degree in a similar discipline without prior industry experience required. Regardless, you must have owned at least one production-quality system (whether industry, research, or a substantial open-source project) from start to finish.

• Proficient in Python engineering. Must demonstrate clean abstractions, type discipline, asynchronous programming, and well-tested code.

• In-depth, practical knowledge of agent loops — understanding how a model decides to utilize a tool, how the tool's results reintegrate into context, how loops conclude, and identifying failure points.

• Hands-on experience with at least one major agent SDK — such as OpenAI Agents SDK, Anthropic SDK / claude-agent-sdk, LangGraph, or a comparable alternative — along with a well-informed opinion on their trade-offs.

• Familiarity with the construction of modern coding agents and their context engineering — including input for the system prompt, file manipulation, planning and tracking long-running tasks, and understanding potential failure points.

• Proficient in using Claude Code / Codex as a power user. You should be capable of brainstorming, planning, and executing complex engineering tasks with these tools — including reading their source code when necessary to comprehend or extend functionality.

• Strong skills in VS Code and git — including branches, rebases, worktrees, conflict resolution, and PR workflows. This is essential.

• A measurement-oriented mindset: you won't ship without an evaluation, and you don't trust a number unless you can replicate it.

🏝️ Benefits

• Work from anywhere in the US! Machinify operates in a digital-first environment.

• Comprehensive Medical/Dental/Vision benefits.

• FSA/HSA options available.

• Tuition reimbursement program.

• Competitive salary along with a 401(k) that includes a company match.

• Unlimited paid time off (PTO).

• Additional health and wellness benefits and perks.

• A flexible and trusting environment where you will feel empowered to perform at your best.