Remote AI Harness Engineer for Local AI & P2P

Tether; København; Remote; DKK 720,000 - 1,320,000; Full time

Job summary

Tether is seeking an AI Harness Engineer to advance P2P and AI tooling, building agent harnesses, tool calling, and scalable loops that enable secure, distributed workflows for a global user base.

Join a remote, globally distributed team, contribute to evals and testing, and craft modular JavaScript/TypeScript code while aligning with robust failure-resilience practices across networks.

Qualifications

• 5+ years building production software, with deep experience in P2P systems, AI/LLM tooling, or both
• Hands‑on experience building agent harnesses, including tool calling, agent loops, context management, and structured outputs
• Experience building evals and test infrastructure for non‑deterministic systems
• Strong JavaScript/TypeScript, with experience modelling complex state, from sync protocols to agent loops
• Open‑source contributions to P2P, networking, or AI tooling projects
• Experience designing for failure, including network partitions, model errors, timeouts, and retries

Responsibilities

• Deliver new features and bug fixes with a focus on reliability, performance, and user satisfaction
• Design and develop new tools and modules, enabling new AI capabilities while ensuring success of a wide variety of model types and sizes
• Maintain high codebase quality through consistent coding standards, code reviews, and automated testing
• Implement and develop P2P technology for use with local AI

Skills

JavaScript/TypeScript

Agent harnesses

Testing & debugging

P2P/distributed systems

Open-source contributions

AI tooling

Tools

npm

Job description

Tether is seeking an AI Harness Engineer to advance P2P and AI tooling, building agent harnesses, tool calling, and scalable loops that enable secure, distributed workflows for a global user base.

Join a remote, globally distributed team, contribute to evals and testing, and craft modular JavaScript/TypeScript code while aligning with robust failure-resilience practices across networks.