Autonomous Driving Multimodal Model Engineer

Black Sesame Technologies Inc; San Jose (CA); On-site; USD 150,000 - 200,000; Full time

Job summary

Black Sesame Technologies Inc in California is seeking an Autonomous Driving Multimodal Model Algorithm Engineer to advance Vision-Language Models, World Models, and BEV-based perception. You'll collaborate across perception, prediction, planning, data, simulation, and deployment teams to integrate multimodal models with BEV perception and end-to-end driving systems within automotive AI hardware.

The role emphasizes hands-on modeling, efficient fine-tuning, and deploying multimodal architectures

Qualifications

• MS or PhD in Computer Science, Electrical Engineering, Robotics, AI, or related field.
• Strong background in deep learning, computer vision, multimodal learning, robotics, or autonomous driving.
• Hands-on experience with Vision-Language Models and multimodal large models.
• Experience with BEV perception, end-to-end autonomous driving, and planning.
• Proficient in PyTorch; familiarity with Python and C++/CUDA is a plus.

Responsibilities

• Develop multimodal models for autonomous driving, including VLM, VLA, and World Models.
• Collaborate across perception, prediction, planning, data, simulation, and deployment teams.
• Adapt open-source multimodal architectures for driving scenarios.
• Align multimodal representations for driving-specific tasks.
• Conduct systematic evaluation and iterate on model deployments.

Skills

Vision-Language Models

Multimodal learning

World models

BEV perception

PyTorch

Python

C++/CUDA

Git

Docker

Linux

Attention mechanisms

Transformer architectures

Education

MS or PhD in CS/EE/Robotics/AI

Tools

LLaVA

Qwen-VL

InternVL

MiniCPM-V

OpenVLA

BLIP-style models

Flamingo-style models

Job description

Black Sesame Technologies Inc in California is seeking an Autonomous Driving Multimodal Model Algorithm Engineer to advance Vision-Language Models, World Models, and BEV-based perception. You'll collaborate across perception, prediction, planning, data, simulation, and deployment teams to integrate multimodal models with BEV perception and end-to-end driving systems within automotive AI hardware.

The role emphasizes hands-on modeling, efficient fine-tuning, and deploying multimodal architectures