Staff VLA Engineer

42dot; Sacramento (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

42dot is a mobility AI company advancing autonomous driving software with a focus on safety and continuous updates. As a Staff VLA Engineer, you will contribute to next-generation VLA architectures and specialize in vision-language-action research, turning ideas into implementable progress.

Your work will span foundation models, multimodal learning, and world models, collaborating across Silicon Valley and HQ Korea to push the boundaries of AI for driving intelligence.

Qualifications

• Strong hands-on experience with deep learning frameworks (e.g., PyTorch, JAX).
• Solid understanding of foundation models, multimodal learning, or transformer-based architectures.
• Experience training or fine-tuning large-scale models (vision, language, or multimodal).
• Strong software engineering skills and experience with large-scale data pipelines.
• Ability to read, implement, and extend ideas from recent AI research papers.

Responsibilities

• Contribute to Next-Generation VLA Architectures.
• Research and prototype next-generation Vision-Language-Action architectures.
• Explore scalable multimodal foundation models for autonomous driving.
• Help develop architectures with improved generalization, reasoning, and long-horizon decision making.
• Support work on large driving models that unify perception, planning, and action generation.
• Support World Model & Driving Reasoning Research; build world-model-based approaches for predictive driving intelligence.
• Work on future-state prediction, behavior forecasting, and counterfactual simulation.
• Contribute to long-horizon planning and reasoning research for autonomous driving.
• Help build models that understand complex traffic interactions and latent agent intentions.
• Explore Agentic Driving Systems; research goal-driven autonomous driving agents.
• Help develop architectures that integrate reasoning, planning, memory, and action.
• Investigate driving agents capable of adaptive decision making in open-world environments.
• Contribute ideas toward future driving-agent architectures as successors to current VLA systems.

Skills

Deep learning frameworks (PyTorch, JAX

Foundation models & multimodal

Large-scale data pipelines

Read/implement AI papers

Cross-functional collaboration

Education

MS or PhD in CS/Robotics/ML/EE

Tools

PyTorch

JAX

Job description

We are looking for the best

About Us

42dot is a mobility AI company committed to solving mobility challenges with software and AI. As the Global Software Center of Hyundai Motor Group, 42dot pioneers the future of mobility by advancing the development of software‑defined vehicles. We develop safety‑first, user‑centric software‑defined vehicle technologies that deliver the latest performance through continuous updates like smartphones. By advancing software and AI technology, 42dot envisions a world where everything is connected and moves autonomously through a self‑managing urban transportation operating system.

As a Staff VLA Engineer you'll contribute to next‑generation autonomous driving intelligence research, working alongside the team to push past current VLA capabilities. You'll bring hands‑on expertise in foundation models, multimodal learning, world models, or autonomous systems, and help turn research ideas into real technical progress.

Responsibilities

• Contribute to Next-Generation VLA Architectures
• Research and prototype next‑generation Vision‑Language‑Action architectures.
• Explore scalable multimodal foundation models for autonomous driving.
• Help develop architectures with improved generalization, reasoning, and long‑horizon decision making.
• Support work on large driving models that unify perception, planning, and action generation.
• Support World Model & Driving Reasoning ResearchBuild and experiment with world‑model‑based approaches for predictive driving intelligence.
• Work on future‑state prediction, behavior forecasting, and counterfactual simulation.
• Contribute to long‑horizon planning and reasoning research for autonomous driving.
• Help build models that understand complex traffic interactions and latent agent intentions.
• Explore Agentic Driving SystemsResearch goal‑driven autonomous driving agents.
• Help develop architectures that integrate reasoning, planning, memory, and action.
• Investigate driving agents capable of adaptive decision making in open‑world environments.
• Contribute ideas toward future driving‑agent architectures as successors to current VLA systems.

Qualifications

• MS or PhD in Computer Science, Robotics, Machine Learning, Electrical Engineering, or a related field (or equivalent practical experience).
• Strong hands‑on experience with deep learning frameworks (e.g., PyTorch, JAX).Solid understanding of foundation models, multimodal learning, or transformer‑based architectures.
• Experience training or fine‑tuning large‑scale models (vision, language, or multimodal).Strong software engineering skills and experience working with large‑scale data pipelines.
• Ability to read, implement, and extend ideas from recent AI research papers.Excellent collaboration and communication skills, comfortable working across distributed, cross‑functional, and cross‑cultural teams (Silicon Valley + HQ Korea).

Preferred Qualifications

• Research experience in autonomous driving, robotics, or embodied AI (e.g., perception, planning, control, or end‑to‑end driving models).Experience with world models, model‑based reinforcement learning, or predictive/generative simulation.Familiarity with Vision‑Language‑Action (VLA) models or agentic AI architectures (reasoning, planning, memory, tool use).
• Publications at top‑tier AI/ML/robotics venues (e.g., NeurIPS, ICML, ICLR, CVPR, CoRL, RSS).Experience with large‑scale distributed training and model optimization.Prior experience contributing research to production systems or bringing prototypes to deployment.
• Familiarity with simulation environments for autonomous driving (e.g., CARLA, nuPlan, Waymo Open Dataset).