WorkBuddy-Agent Harness 算法工程师
Core
Optimize the WorkBuddy Harness agent system's performance and capabilities through research and engineering, focusing on self-evolution, context engineering, multi-agent coordination, and long-horizon tasks.
Role type
Senior Research and Engineering IC (AI Agents)
Builds
Self-evolving agent SDKs, evaluation datasets, and multi-agent collaboration frameworks for real-world business scenarios.
Domain
Artificial Intelligence / Large Language Models / Agent Systems
Deliverable
production ML models
Required skills
LLM and Agent architecture expertise, Rust/Go/C++ programming, system-level optimization, experimental design, technical writing
Preferred skills
Multi-agent coordination, long-term memory mechanisms, context engineering, self-evolution algorithms
Technologies
Transformer, Model Inference, Prefix Cache, Programmatic Tool Calling, Context Compaction, MCP, Agent Memory
Responsibilities
Drive 0-to-1 research exploration and independent problem solving; define evaluation standards and build benchmark datasets; implement agent self-evolution mechanisms and multi-agent collaboration systems; optimize context organization and prompt engineering for complex scenarios.
Seniority
Senior, hands-on IC