This internship is well suited to candidates interested in improving existing methods and developing new techniques for large-scale model training, optimization, and inference, extending models to long-context and long-horizon tasks, and improving inference-time efficiency, reliability, and robustness in high-stakes real-world deployments. They rely on Modal for instant GPU access, sub-second container starts, and native storage, so its simple to serve low-latency inference, fine-tune models, and access production-ready sandboxes at scale.