智能体- Agent Harness 全栈工程师- WorkBuddy/CodeBuddy
Core
Full-stack development and architecture of the Agent Harness AI infrastructure to support core capabilities for multiple upper-layer products.
Role type
Senior Full-Stack Engineer (AI Infrastructure)
Builds
End-to-end AI infrastructure including developer tools, visualization interfaces, execution trace replay, context orchestrators, tool registries, and evaluation data management.
Domain
Artificial Intelligence / Large Language Models / Agent Systems
Deliverable
production ML models
Required skills
TypeScript/JavaScript, Python, Go, React/Vue, Node.js, PostgreSQL, Redis, Transformer architecture, model inference, function calling, context window management, MCP, agent memory, multi-agent coordination
Preferred skills
prefix cache optimization, streaming output, task queues, sandbox execution, observability, AI developer tool interaction patterns
Responsibilities
Design and implement end-to-end AI infrastructure from frontend to deployment; build developer tools and visualization interfaces; collaborate with algorithm teams to engineer research results; optimize runtime and architecture for stability and efficiency; iterate on developer experience based on user feedback.
Seniority
Senior, hands-on IC
