Developer Platform Engineer, Agentic (Mid/Senior/Principal)

Positron AI is hiring a Developer Platform Engineer to build the interfaces, workflows, and guardrails that enable AI coding agents to work reliably on its simulators. The role focuses on automating simulator development and model bring-up, verifying agent-produced changes, and measuring agent performance.

Responsibilities:

• Design and build the MCP servers and APIs that let AI coding agents drive our simulators end to end, so that running a simulation, reading a trace, and comparing against a reference all happen from one call
• Make the simulators legible to agents and people alike through clear CLI and API design, structured outputs, reproducible runs, and useful error messages
• Build agent workflows for simulator development in which an agent implements or extends a hardware command's functional, dependency, cost, and cycle models and lands it with test evidence
• Partner with the model architecture engineer to automate new-model bring-up from model specification to simulator coverage
• Build CI guardrails for agent-produced changes, including golden-reference generation, static analysis, known-good comparisons, and provenance on every result
• Measure what agents get right and wrong, and feed those findings back into the tooling

Requirements:

• 4+ years of experience in developer tooling, developer platforms, or infrastructure engineering
• Demonstrated experience building tooling that LLM coding agents use in production, such as MCP servers or tool APIs, agent workflows in CI, evaluation harnesses, or agent-driven code generation with verification
• A strong understanding of CI/CD and test infrastructure, including flaky-test detection, golden-reference testing, and static analysis in CI
• Hands-on experience designing CLIs and APIs that are easy to use correctly and hard to use wrongly, with structured logging and reproducible builds
• Ability to design systems in which agent-produced changes carry evidence and cannot land unverified
• Strong Rust skills, or expert-level C/C++ skills and a willingness to work in Rust day to day, along with proficiency in Python for tooling
• Excellent communication skills, with the ability to explain to engineers what the agent layer can and cannot be trusted with, and why
• This position is open to candidates currently authorized to work in the U.S. We cannot provide new visa sponsorship for this role but are open to facilitating H-1B visa transfers for eligible candidates
• A background in simulation, compilers, or hardware
• Experience running evaluations of agent or model performance on engineering tasks