warmwind

As a Senior Machine Learning Engineer at warmwind you will drive next-gen AI models, working with massive compute clusters (500+ H100 GPUs) and advanced reinforcement learning techniques. Responsibilities include designing, training, and optimizing Large Language Models (LLMs) from scratch; scaling distributed training on large GPU clusters; implementing advanced RL techniques (RLHF, adversarial self-play, real-time control); developing high-performance architectures for multi-modal AI systems; building simulation environments for RL-based agents; and optimizing inference speed and efficiency for deployment. Profile: deep expertise in LLMs and experience building/training large-scale models, experience with distributed training on 500+ GPU superclusters, strong understanding of reinforcement learning and neural network optimization, expertise in PyTorch, TensorFlow, JAX and low-level optimization (CUDA, Triton, DeepSpeed), familiarity with HPC technologies (NVLink, InfiniBand, parallel computing). Strong publication record is a plus. Relocation to Jena, Germany after initial onboarding. Language: German or English.