Job Description
📋 Description develop latent world models for environment dynamics with imagined rollouts (Dreamer-style architect action/policy pipelines (vision-language-action, diffusion-based learning) build generative simulators for controllable future states (video world models) develop multimodal generative capabilities for reasoning and decision-making lead cross-team decisions on training frameworks, data pipelines & evaluation infra drive research-to-production pathways for reliable platform capabilities 🎯 Requirements 8+ years in ML engineering or AI research (or equivalent) deep expertise in at least two of: representation learning, world models, RL, generative modeling hands-on experience training large-scale vision/language/multimodal models with distributed compute strong software engineering fundamentals: system design, performance optimization, production-grade ability to drive cross-team initiatives with ambiguity track record translating research into scalable systems 🎁 Benefits bonus points for latent dynamics modeling, model-based RL, physics-informed NN contributions to open-source ML frameworks or foundation model training codebases background in scientific/structured models (molecular, materials, weather) experience building controllable video generation or neural simulation environments top venues publications (NeurIPS, ICML, ICLR, CVPR, CoRL, RSS)
