[Skip to main content](#main-content) [OpenNash Jobs](https://jobs.opennash.com/) [Hiring report](/reports/hiring/) · [All Tencent Holdings jobs](https://jobs.opennash.com/companies/tencent-holdings/) · [Search jobs](/search/) Tencent Holdings · Chengdu # 企业微信-大模型算法工程师-Agent后训练（北京/广州） Chengdu Listed on Tencent Holdings’s official careers site. Listed as of October 19, 2026. Check that it is still open and apply there. [Apply on Tencent Holdings’s careers site →](https://careers.tencent.com/en-us/jobdesc.html?postId=2094682199537401856) [OpenNash Jobs](https://jobs.opennash.com/) / [Tencent Holdings](https://jobs.opennash.com/companies/tencent-holdings/) / 企业微信-大模型算法工程师-Agent后训练（北京/广州） ## Job description ### Responsibilities - 参与 Agent 大模型的后训练工作，在 SFT、奖励模型、强化学习等方向进行方案设计、数据构建与训练落地，持续提升真实场景下的规划、推理、工具调用与指令遵循能力； - 针对 Agent 任务执行中的能力短板（如多步规划、抗干扰与幻觉抑制等），设计定向的训练优化方案，并构建数据合成、清洗与质量评估管线；面向意图识别等特定场景模型进行训练与优化，结合业务需求完成定制化的能力构建； - 持续跟踪并引入后训练领域的最新技术，推动技术在实际业务的创新落地。 Requirements - 计算机、人工智能、机器学习、数学等相关专业，硕士及以上学历，具备扎实的计算机基础与算法基础，2 年以上相关领域工作经验； - 深入理解大语言模型与 Agent 技术体系，熟练掌握主流后训练技术（SFT、RLHF、DPO、PPO/GRPO 等），具备完整的模型训练与优化项目经验，能独立定位并解决效果问题； - 熟练掌握 Python，熟练使用主流深度学习与训练框架（如 PyTorch、DeepSpeed/Megatron、veRL/OpenRLHF 等），具备扎实的算法实现与大规模分布式训练调优能力； - 熟悉数据构建与评估体系，有强化学习环境搭建、奖励模型训练或高质量指令/偏好数据合成经验者优先； - 在 ICLR、NeurIPS、ICML、ACL、EMNLP 等顶尖会议发表过论文，或为优秀机器学习开源项目贡献者优先。 ## At a glance - Posted by the employer: Not stated - First seen by OpenNash: Oct 12, 2026 - Last checked: Oct 19, 2026 - Listed for: 7 days - Location: Chengdu[Apply on Tencent Holdings’s careers site →](https://careers.tencent.com/en-us/jobdesc.html?postId=2094682199537401856) ## More jobs at Tencent Holdings ### [AI 原生多模数据库内核研发工程师](https://jobs.opennash.com/companies/tencent-holdings/jobs/eb69abf322a31c9b653e/) Chengdu ### [AI 原生多模数据库管控研发工程师](https://jobs.opennash.com/companies/tencent-holdings/jobs/b647a6aa0a4ef45f2732/) Chengdu ### [《QQ三国》游戏运营](https://jobs.opennash.com/companies/tencent-holdings/jobs/d6fb502cd98698b37e05/) Chengdu ### [《王者万象棋》客户端高级开发工程师-包体优化方向](https://jobs.opennash.com/companies/tencent-holdings/jobs/c702e76e636622fa99b0/) Chengdu ### [《王者万象棋》服务器开发工程师](https://jobs.opennash.com/companies/tencent-holdings/jobs/e1731d4ae968af53fa5f/) Chengdu ## Similar jobs at other companies ### [大模型算法工程师-实习](https://jobs.opennash.com/companies/xiaomi/jobs/213a1b9f5784b73bb705/) Xiaomi · 武汉 ### [大模型算法工程师](https://jobs.opennash.com/companies/midea-group/jobs/668f44517e0f3c6ddbe2/) Midea Group · 佛山市 +1 more ### [大模型算法工程师-Airc](https://jobs.opennash.com/companies/midea-group/jobs/5dd2365ab68184fd57df/) Midea Group · 上海市 ### [大模型算法工程师-运筹方向](https://jobs.opennash.com/companies/sf-holding/jobs/55bec1d5db69b929731c/) S.F. Holding · 广东省/深圳市 ### [大模型算法工程师-群体智能方向](https://jobs.opennash.com/companies/sf-holding/jobs/15895c0f9da86b7c8c12/) S.F. Holding · 广东省/深圳市 OpenNash lists jobs from employers’ own careers sites and does not handle applications. Dates come from the employer where stated; “first seen” and “last checked” are our own observations. OpenNash Jobs · Official listings with dates and links to the employer. [Weekly hiring report](/reports/hiring/) · [Methodology](/methodology/)