VLA Pre-training Lead (Deep Learning)

Join Humanoid, a company on a mission to build the world's most capable and safe humanoid robots. As the VLA Pre-training Lead, you will own the pretraining stage of our VLM and VLA-based framework, define the base model strategy, and lead a team of research engineers. You will also design and run large-scale distributed training, define pretraining-scale data, and establish continuous pretraining pipelines. This role requires a track record of building deep-learning systems, experience pretraining large models, and a deep understanding of transformer and diffusion architectures.

• Own the VLA pretraining roadmap end-to-end, including architecture choices, data mixtures, scaling laws, and evaluation protocols for base models.
• Lead, grow, and mentor a team of deep learning engineers focused on pretraining, setting research direction and engineering standards.
• Design and run large-scale distributed training on multi-node GPU clusters; drive throughput, stability, and cost efficiency in partnership with MLOps & Data Platform teams.