PhD AI Research Intern: World Models for Autonomous Driving

Socket.dev; Santa Clara (CA); On-site; USD 55,000 - 83,000; Full time

Benefits offered by this job

Fun, supportive environment

Computational resources

Cutting-edge projects

Impact on transportation revolution

Competitive compensation

Snacks, lunches, dinners

Job summary

XPENG is seeking PhD research interns with expertise in generative modeling to develop world models for autonomous driving and robotics. You will work with our research team on diffusion/flow-matching architectures using large-scale multimodal data from our fleet.

Strong publication potential is encouraged and supported. Work will be conducted on-site at our Santa Clara office, with mentorship and opportunities to publish results and open-source code.

Qualifications

• Pursuing a PhD in Engineering, Computer Science, or related field with focus on DL/CV/Generative Models.
• First-author publications at top-tier venues (CVPR, ICCV, ECCV, NeurIPS, ICLR, ICML, CoRL, RSS, or SIGGRAPH).
• Strong foundation in generative modeling and experimental methodology with hands-on experience in PyTorch or JAX.
• Strong Python programming and software design skills for large-scale data processing.

Responsibilities

• Drive a focused research project on predictive world models end to end with mentor and team.
• Contribute to multi-view prediction and action-conditioned forecasting architectures or unsupervised forecasting.
• Extend predictions into a 3D multimodal latent space for simulation and evaluation.
• Investigate cross-embodiment generalization across vehicles and sensor configurations.
• Build evaluation methodologies for model quality, accuracy, fidelity, and policy performance.
• Collaborate with engineers to move prototypes to scalable training and publish results.

Skills

Deep Learning

Computer Vision

Generative Models

PyTorch

JAX

Education

PhD in Engineering/CS or related

Tools

PyTorch

JAX

Job description

XPENG is seeking PhD research interns with expertise in generative modeling to develop world models for autonomous driving and robotics. You will work with our research team on diffusion/flow-matching architectures using large-scale multimodal data from our fleet.

Strong publication potential is encouraged and supported. Work will be conducted on-site at our Santa Clara office, with mentorship and opportunities to publish results and open-source code.