*CodSpeed is the continuous performance optimization platform trusted by world-class engineering teams. We shift performance testing left to catch regressions before they ship. From open-source projects like Mozilla, Pydantic, and Astral to enterprise teams at Cloudflare, Vercel, and beyond. CodSpeed empowers developers to measure, track, and improve software efficiency across the stack.*

We're a small, deeply technical team looking for engineers who want real ownership, direct user contact, and the chance to ship features end-to-end.

Tech stack

Technologies: Python · FastAPI · Rust· Docker

Tooling: GitHub · Linear · Claude Code

What you'll do

• Own the AI/agent layer: connect our instruments to agents & build on top of these
• Design evaluation and QA methodology for agentic coding features
• Work on context, data pipelines, model selection, and iteration loops
• Mine our existing dataset to extract performance insights and patterns

Requirements

• Strong engineering fundamentals: you ship fast without sacrificing quality
• Pragmatic, product-oriented mindset: not research for research's sake
• Strong Python and production engineering habits (FastAPI, typed Python, services, testing)
• Ability to design evaluation and quality control methodology
• High autonomy: can scope, prioritize, and deliver independently
• Comfortable working in an early-stage environment
• 5+ years of experience as a technical contributor (professional or open source)

Extras

• Experience with LLM agents, eval frameworks, or scoring systems
• Sensitivity to performance and measurement (important for our DNA)