AI Agent研发工程师（模型评测方向） - 移动OS
Core
Build AI agent infrastructure, evaluation frameworks, and self-evolving loops for mobile OS smart hardware interactions.
Role type
Senior IC AI Agent Research Engineer (Model Evaluation & Infrastructure)
Builds
Agent Infra, evaluation systems, AgenticSearch, knowledge bases, and self-evolving feedback loops for mobile OS.
Domain
Mobile OS / AI Agents / Large Language Models
Deliverable
production ML models
Required skills
Python, Go/Nodejs/TypeScript, system design, LLM API integration, Prompt engineering, Agent Loop/ReAct/Plan-and-Execute architectures, trajectory analysis, context engineering.
Preferred skills
Agent evaluation体系建设, LLM fine-tuning, data flywheel design, Multi-Agent orchestration, RAG/vector retrieval, MCP/Function Calling, AI coding tools.
Technologies
Python, Go, Nodejs, TypeScript, LLM APIs, Vector Search, Knowledge Graphs, MCP, Function Calling.
Responsibilities
Build and optimize Agent Infra and evaluation frameworks; construct multi-dimensional evaluation strategies (deterministic, model-based, human-in-the-loop); design self-evolving feedback loops for capability updates; track AI trends to drive product innovation.
Seniority
Senior, hands-on IC
