AI Harness Engineer in New York at Jobgether

This position is listed on behalf of a partner company, who manages all applications and next steps. Our partner is looking for an AI Harness Engineer based in United States.

As an AI Harness Engineer, you will build the systems and integrations that connect people with powerful local AI capabilities. You’ll work at the intersection of AI tooling, agent infrastructure, and peer-to-peer technology, helping create scalable experiences for the next generation of AI assistants. Your work will include developing reusable tools and modules, supporting diverse AI model types and sizes, and building reliable systems for non-deterministic environments. You’ll contribute to the evolution of P2P technology designed to enable local AI at scale. The role combines deep software engineering with experimentation and rapid technical exploration. You’ll join a remote, globally distributed environment where autonomy, technical curiosity, and continuous innovation are highly valued.

• Build and deliver new features and bug fixes with a strong focus on reliability, performance, and user satisfaction.
• Design and develop reusable tools, modules, and integrations that enable new AI capabilities across different model types and sizes.
• Build and maintain AI agent harnesses, including tool calling, agent loops, context management, and structured outputs.
• Develop evaluation frameworks and testing infrastructure for AI systems with non-deterministic behavior.
• Implement and evolve peer-to-peer technology to support local AI and distributed use cases.
• Maintain a high-quality codebase through consistent coding standards, code reviews, modular architecture, and automated testing.
• Design systems that can reliably handle failures such as network partitions, model errors, timeouts, and retries.
• Develop and maintain reusable npm modules and shared software components.
• Experiment with new technologies and approaches to continuously expand AI and P2P capabilities.
• Collaborate effectively with a globally distributed engineering team while communicating technical ideas clearly.

• 5+ years of experience building and maintaining production software.
• Strong proficiency in JavaScript and TypeScript, including experience modeling complex state across synchronization protocols, agent loops, or similar systems.
• Strong ability to write modular, maintainable, and reusable code.
• Hands-on experience building AI agent harnesses, including tool calling, context management, agent loops, and structured outputs.
• Experience developing evaluation systems and test infrastructure for non-deterministic AI or other complex systems.
• Experience creating and managing reusable npm modules.
• Strong testing and debugging skills, with a focus on reliability and production quality.
• Strong understanding of and practical experience developing and using AI tools.
• Deep interest or professional experience with peer-to-peer technologies, networking, or distributed systems.
• Experience designing resilient systems that account for network failures, model errors, timeouts, retries, and other failure scenarios.
• Contributions to open-source projects related to P2P, networking, AI tooling, or adjacent technologies are valued.
• Strong communication skills and the ability to work effectively in a remote, globally distributed environment.
• A highly experimental, hands-on mindset with a willingness to continuously test ideas, explore emerging technologies, and solve challenging technical problems.

• Fully remote work environment.
• Opportunity to work on emerging AI, agent infrastructure, and peer-to-peer technologies.
• Exposure to complex distributed systems and non-deterministic AI applications.
• Opportunity to contribute to open-source P2P, networking, and AI tooling ecosystems.
• International and globally distributed engineering environment.
• High degree of autonomy and room for experimentation and technical innovation.