Calibre - Product Engineer
Requirements
• Demonstrable AI engineering depth. You can show real systems where you made substantive decisions about agent harnesses, evals, hallucination containment, performance, agent sandboxing or MCP/tool design. • Demonstrable AI engineering depth. • Strong product engineering and technical judgment. You ship across interface, backend, data and deployment, reason from first principles and choose tools deliberately. • Strong product engineering and technical judgment. • Product instinct and clear collaboration. You ask good questions, form a view, test quickly, communicate clearly and care whether the work changes outcomes. • Product instinct and clear collaboration. • Evidence of exceptional output. Show us professional work, open-source contributions, personal products, research or hackathon results. Roughly 2+ years of experience is useful, but evidence matters more than tenure. • Evidence of exceptional output. • Exceptional technical talent with more than 1 year of professional experience. Your GitHub, personal projects, or hackathon wins speak for themselves. • Exceptional technical talent • A first-principles problem solver. You have a fundamentally strong computer science background (e.g., from a top university or equivalent experience) and are language-agnostic, picking the right tool for the job. • first-principles problem solver. • AI-Native: You are already a sophisticated user of LLM tools (e.g., Claude Code, Cursor, etc.) for coding. You understand their strengths, weaknesses, and how to build systems that leverage them effectively. • AI-Native: • High-Leverage & High-Output: You thrive in a high-pressure, high-growth environment. You understand that in an early-stage startup, precision and speed are paramount. • High-Leverage & High-Output:
Responsibilities
• Own products across the whole stack. Build and ship high-quality, high-velocity code across the full stack (front-end, back-end, and AI-agent infrastructure) • Own products across the whole stack. • Build products end-to-end. Design systems from discovery through to production. • Engineer reliable agent systems. Design harnesses, evals and safeguards that make quality non-compromisable.
Benefits
• £60K – £150K • 0.25% – 1% • AI systems are moving from proof of concepts into the infrastructure of everyday life. That creates a fundamental technical problem: how do you know an increasingly autonomous AI system is secure, reliable and behaving as intended? And who do you trust to answer that question? • The institutions and processes we rely on to provide that confidence were built for a slower, more predictable world. AI is advancing faster than the mechanisms we use to trust it. • Calibre is building trust for the AI era as the first AI-native certification body. We develop core technologies to understand AI and cybersecurity risk, and use them to independently assess companies against leading standards. • Founded by two ex-Palantir technical co-founders, and a founding team spanning audit and AI research, we are pushing the frontier of what trust looks like for the AI economy. We’ve raised $3.3M from top tier SF based VCs.
Apply in one click
Upload My Resume
Drop here or click to browse · Tap to choose · PDF, DOCX, DOC, RTF, TXT