1. Develop VLA/VLM models based on image and perceptual information to enhance the model's ability to understand and execute vision-language-action tasks in complex scenarios;
2. Lead research on cutting-edge algorithms for VLA/VLM models and drive optimization efforts, spearheading data construction and task-driven modeling;
3. Collaborate closely with engineering teams to advance the practical implementation of algorithms in real-world systems such as robots and drones.

1. Master's degree or higher in Computer Science, Electronic Engineering, Applied Mathematics, Automation, or related fields;
2. Understanding of VLA/VLM model training processes and deployment, with familiarity in mainstream deep learning frameworks;
3. Background in at least one of the following algorithms: large language models, multimodal large models, generative policies, or world models; Preferred Skills:
a. Familiarity with end-to-end visual navigation and motion planning algorithms is preferred;
b. Experience in building complete datasets is preferred;
c. Experience in ACM competitions is preferred;
d. Experience in AI algorithm R&D for autonomous driving or robotics is preferred.