智能体算法专家（J104318）
Core
Design and optimize core algorithms for general-purpose AI agents, enabling end-to-end task execution from understanding to delivery.
Role type
Senior IC Agent Algorithm Engineer
Builds
General-purpose AI agents capable of multi-step task planning, tool use, and state management
Domain
Artificial Intelligence / Large Language Models / Agentic Systems
Deliverable
production ML models
Required skills
Large Language Models, ReAct, Planning, Tool Use, Memory, Reflection, RAG, Python, Algorithm Engineering, Model Evaluation, Experiment Design
Preferred skills
Agentic RL, Tool-use Training, Trajectory Data, Process Reward, AgentScope, LangGraph, OpenAI Agents SDK, AutoGen, Distributed Task Scheduling, State Machines, Redis, Message Queues
Responsibilities
Design and optimize agent core algorithms for task understanding, planning, decision-making, and tool execution; Build capabilities for tool/skill recall, selection, routing, and parameter generation; Develop context and state management systems including working memory and checkpointing; Establish evaluation and attribution systems based on Agent Trace; Collaborate with model, search, and infrastructure teams to drive online deployment.
Seniority
Senior, hands-on IC
