AGIBOT

具身算法工程师(VLA/WAM/RL方向)

AGIBOT  •  Onsite  •  3 hours ago
Apply
AI can make mistakes so check important info. Chat history is never stored.

Job Description

具身算法工程师(VLA/WAM/RL方向)

上海

全职

职位描述

1、参与机器人数据采集、算法设计、模型训练与真机部署的全链路研发工作; 2、参与机器人数据采集规划,包括任务设计与质量分析,为下游模型训练提供高质量、多样化数据支持; 3、跟踪具身智能、机器人学习、VLA/WM/WAM 等方向的前沿进展,调研并分析最新研究工作。负责算法创新、模型结构设计与训练优化; 4、探索 RL 与模仿学习(IL)的混合训练范式:利用人类演示数据初始化策略,再通过 RL 进行微调和鲁棒性增强; 5、设计模型与底层运控(WBC/MPC)的协同接口: 实现高层 planner 策略输出与低层wbc controller的平滑衔接,确保物理安全性与稳定性; 6、参与模型在真实机器人平台上的模型部署与实时性优化、调试与性能评估,推动算法从仿真/离线训练走向真实场景应用;

职位要求

1、具备扎实的编程能力和工程实现能力,熟悉 PyTorch 等深度学习框架。熟练使用 Isaac Sim、MuJoCo 等仿真环境。熟悉机器人学基础:熟悉运动学(FK/IK)、动力学、坐标系变换(SMPL Y-up/Z-up 等)、关节空间与任务空间映射、全身控制(WBC)概念。 2、对深度学习基础模型结构、模仿学习方法有较好的理解。掌握以下基础内容: - 基本深度学习知识,如Transformer, DiT等模型架构。 - 基本现代生成模型知识,如Diffusion Model, Flow-Based Model等。 - 基本模仿学习知识,如ACT, Diffusion Policy等 3、熟悉具身智能算法方向的前沿进展,了解VLA、WAM、机器人基座模型的最新相关工作。尤其是: - 主流的VLA模型,如Pi系列,gr00t系列等。 - 主流的WAM模型, 如DreamZero,Lingbot-VA, FastWAM等。

投递

AGIBOT

About AGIBOT

AGIBOT is an Embodied AI foundation model company developing both the intelligence layer and the corresponding robotic embodiments needed to bring general intelligence into the physical world. Founded in 2023, AGIBOT is building a full-stack Embodied AI platform across foundation models, data infrastructure, simulation, robotic hardware, and deployment. AGIBOT’s “Three Intelligences in One” architecture integrates Locomotion Intelligence, Interaction Intelligence, and Manipulation Intelligence into a unified embodied system. AGIBOT’s portfolio spans humanoid robots, quadrupeds, dexterous systems, and commercial cleaning solutions. In March 2026, AGIBOT announced that its 10,000th robot had rolled off the production line.

Industry
Manufacturing & Production
Company Size
201-500 employees
Headquarters
Shanghai, CN
Year Founded
Unknown
Social Media