Harness Engineer, Product

About the role

This role is responsible for the glue between inference pipelines, vector stores, workflow engines, and the UI. You'll design how AI capabilities are composed, triggered, and surfaced to end users across the full stack. You will deliver harnesses that allow agents to perform real-world work.

What you'll do

• Build and maintain the AI harness layer that orchestrates models, tools, and data pipelines
• Design and implement APIs that connect frontend interfaces to inference and retrieval systems
• Develop agent workflows and tool-use patterns that chain multiple AI capabilities together
• Optimize end-to-end latency from user input through model inference to rendered response
• Collaborate with frontend and backend engineers to ship integrated AI features
• Write tests and documentation for critical orchestration and pipeline paths

Qualifications

• Applicant MUST be a U.S. citizen and reside in the U.S. while performing work
• 3+ years of professional software engineering experience
• Strong proficiency in TypeScript, React and web tech (web-sockets, SSE, https)
• Experience building orchestration layers or middleware that connect multiple services
• Familiarity with vector databases (e.g., Pinecone, Weaviate, ChromaDB)
• Understanding of agent frameworks, tool-use patterns, and retrieval-augmented generation
• Must work collaboratively with others who possess diverse skillsets and beliefs

Nice to have

• Have built apps with Tauri, Node.js or Electron
• Experience designing workflow or pipeline orchestration systems
• Familiarity with llama.cpp, vLLM, Transformers, ONNX or other local inference runtimes
• Experience with Python for ML/AI workloads or Natural Language Processing techniques
• Experience in high security and compliance industries (Healthcare, Government, Defense)

Benefits

• Competitive salary
• Flexible hours
• Training and professional development