小鹏汽车Research Scientist (Embodied AI)
任职要求
1、计算机、电子工程、人工智能等相关领域硕士及以上学历； 2、具有扎实的机器学习算法基础，在VLM、LLM、AIGC、RL、Robotics、3D Vision等相关专业领域有研究经验，曾以第一作者身份在CVPR/ECCV/ICCV/CoRL/ICRA/NeurIPS/ICLR/ICML/SIGGRAPH等顶会顶刊上发表过论文； 3、熟练使用PyTorch/TensorFlow等深度学习框架，具备良好的代码实现能力； 4、具有良好的团队合作能力和沟通能力。 加分项 1、计算机、电子工程、人工智能、机器人等相关领域博士学历； 2、有多模态、大模型、机器人相关研究和项目经验，有国际影响力的论文主要作者或项目主导者； 3、具有优秀的代码能力，如ACM/ICPC、NOI/IOl、Top Coder、Kaggle等比赛获奖者； 4、具备解决复杂问题的经验，并能比较各种解决方案，并根据不同的视角确定前进方向。具备基于机器学习和/或深度学习方法构建系统的经验。
工作职责
1、面向通用人形机器人，打造行业领先的具身大模型，实现全身动作控制、场景及任务可泛化的、可scaling的VLA大模型研究，形成持续的技术影响力并引领国际行业发展。
1、面向通用人形机器人，打造行业领先的具身大模型，实现全身动作控制、场景及任务可泛化的、可scaling的VLA大模型研究，形成持续的技术影响力并引领国际行业发展。
· Lead strategic analysis on global trends in Physical AI, including autonomous driving, embodied intelligence, robotics, and sensorimotor foundation models. · Continuously monitor and benchmark key players (e.g., Tesla FSD, Waymo Robotaxi) and emerging technologies in North America and globally. · Conduct exploratory research on cutting-edge technologies related to world models, multimodal large models, and planning frameworks in AI. · Produce high-quality technical and strategic insight reports for internal stakeholders to inform roadmap and business decisions. · Collaborate with R&D and product teams to shape future research directions and support decision-making with clear, data-driven foresight. · Participate in academic engagements, including paper drafting and submission to top-tier AI conferences (e.g., NeurIPS, CVPR, ICRA, RSS).
• Lead the development and evaluation of state-of-the-art models for code completion and editing, pushing the boundaries of code understanding, generation, fix and review. • Develop retrieval-augmented systems that improve a model’s awareness of large and complex codebases, enabling context-rich code assistance. • Design and prototype efficient inference algorithms to enable fast, interactive code generation experiences at • Collaborate across disciplines with product teams across Microsoft and Github • Stay up to date with the research literature and product advances in AI for software engineering