企业微信-Agent算法工程师-（广州/北京）
Tencent
企业微信-Agent算法工程师-（广州/北京）
Job Description
企业微信-Agent算法工程师-（广州/北京）
📋 Job Overview
The Enterprise WeChat Agent Algorithm Engineer role involves developing advanced multi-agent systems that enable autonomous decision-making, collaboration, and tool invocation to transition large models from conversational AI to task-executing agents. Responsibilities include researching and applying complex reasoning techniques like Chain of Thought (CoT) and Tree of Thought (ToT) to tackle open-ended problems, and using methods such as instruction fine-tuning and reinforcement learning to enhance model planning, reasoning, and instruction adherence. This position is based in Guangzhou or Beijing and focuses on solving automation challenges in complex scenarios.
📍 Location: Chengdu, China
🏢 Business Unit: WXG
📄 Full Description
2.深入研究并应用LLM的复杂推理技术（如思维链CoT、思维树ToT），攻克开放式、复杂问题的深度研究模式，赋予Agent独立探索与解决问题的能力；
3.运用指令微调、强化学习等方法，提升大模型规划、推理与遵循指令的能力，从而提升模型知识边界探索和抗干扰的能力。
🎯 Key Responsibilities
- 负责构建具备自主决策、协作与工具调用能力的多智能体系统（Multi-Agent System），推动大模型从“对话式AI”向“任务执行体”演进，解决复杂场景下的自动化问题
- 深入研究并应用LLM的复杂推理技术（如思维链CoT、思维树ToT），攻克开放式、复杂问题的深度研究模式，赋予Agent独立探索与解决问题的能力
- 运用指令微调、强化学习等方法，提升大模型规划、推理与遵循指令的能力，从而提升模型知识边界探索和抗干扰的能力
🛠️ Required Skills
- Multi-Agent System development
- LLM complex reasoning techniques (CoT, ToT)
- Instruction fine-tuning
- Reinforcement learning
- Model planning and inference enhancement
Locations
- Chengdu, China
Salary
300,000 - 600,000 CNY / yearly
Source: ai estimated
* This is an estimated range based on market data and may vary based on experience and qualifications.
Skills Required
- Multi-Agent System developmentintermediate
- LLM complex reasoning techniques (CoT, ToT)intermediate
- Instruction fine-tuningintermediate
- Reinforcement learningintermediate
- Model planning and inference enhancementintermediate
Responsibilities
- 负责构建具备自主决策、协作与工具调用能力的多智能体系统（Multi-Agent System），推动大模型从“对话式AI”向“任务执行体”演进，解决复杂场景下的自动化问题
- 深入研究并应用LLM的复杂推理技术（如思维链CoT、思维树ToT），攻克开放式、复杂问题的深度研究模式，赋予Agent独立探索与解决问题的能力
- 运用指令微调、强化学习等方法，提升大模型规划、推理与遵循指令的能力，从而提升模型知识边界探索和抗干扰的能力
Tags & Categories
Answer 10 quick questions to check your fit for 企业微信-Agent算法工程师-（广州/北京） @ Tencent.
Related Books and Jobs
企业微信-Agent算法工程师-（广州/北京）
Tencent
企业微信-Agent算法工程师-（广州/北京）
Job Description
企业微信-Agent算法工程师-（广州/北京）
📋 Job Overview
The Enterprise WeChat Agent Algorithm Engineer role involves developing advanced multi-agent systems that enable autonomous decision-making, collaboration, and tool invocation to transition large models from conversational AI to task-executing agents. Responsibilities include researching and applying complex reasoning techniques like Chain of Thought (CoT) and Tree of Thought (ToT) to tackle open-ended problems, and using methods such as instruction fine-tuning and reinforcement learning to enhance model planning, reasoning, and instruction adherence. This position is based in Guangzhou or Beijing and focuses on solving automation challenges in complex scenarios.
📍 Location: Chengdu, China
🏢 Business Unit: WXG
📄 Full Description
2.深入研究并应用LLM的复杂推理技术（如思维链CoT、思维树ToT），攻克开放式、复杂问题的深度研究模式，赋予Agent独立探索与解决问题的能力；
3.运用指令微调、强化学习等方法，提升大模型规划、推理与遵循指令的能力，从而提升模型知识边界探索和抗干扰的能力。
🎯 Key Responsibilities
- 负责构建具备自主决策、协作与工具调用能力的多智能体系统（Multi-Agent System），推动大模型从“对话式AI”向“任务执行体”演进，解决复杂场景下的自动化问题
- 深入研究并应用LLM的复杂推理技术（如思维链CoT、思维树ToT），攻克开放式、复杂问题的深度研究模式，赋予Agent独立探索与解决问题的能力
- 运用指令微调、强化学习等方法，提升大模型规划、推理与遵循指令的能力，从而提升模型知识边界探索和抗干扰的能力
🛠️ Required Skills
- Multi-Agent System development
- LLM complex reasoning techniques (CoT, ToT)
- Instruction fine-tuning
- Reinforcement learning
- Model planning and inference enhancement
Locations
- Chengdu, China
Salary
300,000 - 600,000 CNY / yearly
Source: ai estimated
* This is an estimated range based on market data and may vary based on experience and qualifications.
Skills Required
- Multi-Agent System developmentintermediate
- LLM complex reasoning techniques (CoT, ToT)intermediate
- Instruction fine-tuningintermediate
- Reinforcement learningintermediate
- Model planning and inference enhancementintermediate
Responsibilities
- 负责构建具备自主决策、协作与工具调用能力的多智能体系统（Multi-Agent System），推动大模型从“对话式AI”向“任务执行体”演进，解决复杂场景下的自动化问题
- 深入研究并应用LLM的复杂推理技术（如思维链CoT、思维树ToT），攻克开放式、复杂问题的深度研究模式，赋予Agent独立探索与解决问题的能力
- 运用指令微调、强化学习等方法，提升大模型规划、推理与遵循指令的能力，从而提升模型知识边界探索和抗干扰的能力
Tags & Categories
Answer 10 quick questions to check your fit for 企业微信-Agent算法工程师-（广州/北京） @ Tencent.
Related Books and Jobs