The company is looking to contribute to the training and optimization of foundational generative models, including LLMs, diffusion models, and multimodal architectures, and build the core infrastructure, tools, and pipelines for data preprocessing, model training, evaluation, and deployment.
Requirements
- Proficiency in Python and deep learning frameworks such as PyTorch or JAX
- Experience in at least one of the following areas: Large-scale model training and optimization (LLMs, diffusion, transformers, etc.)
- Experience in at least one of the following areas: Scalable infrastructure for ML training (e.g., Ray, Kubernetes, distributed systems)
- Experience in at least one of the following areas: Efficient data pipelines for training foundational models on multi-modal data (text, image, video, 3D)
- Experience in at least one of the following areas: Systems-level contributions for compute, parallelism (e.g., FSDP, ZeRO, sharding), or inference efficiency
- Familiarity with evaluating generative models, including metrics, benchmarks, and human-in-the-loop evaluation
- Publication record in top-tier ML or systems conferences (e.g., NeurIPS, ICML, ICLR, CVPR, SOSP, MLSys) is a plus
Responsibilities
- Contribute to the training and optimization of foundational generative models, including LLMs, diffusion models, and multimodal architectures (e.g., video, 3D, audio)
- Design scalable algorithms and training strategies for large-scale distributed training (multi-GPU, multi-node)
- Build core infrastructure, tools, and pipelines for data preprocessing, model training, evaluation, and deployment
- Explore and integrate cutting-edge techniques in model compression, parallelism, and efficient inference
- Collaborate across research, engineering, and product teams to align system design with product applications and deployment goals
- Stay up to date with state-of-the-art advancements in generative modeling, training infrastructure, and optimization — and publish when applicable
Other
- Advanced degree in Computer Science, Machine Learning, or related field (or equivalent practical experience in training large models)
- Strong collaboration and communication skills; ability to work across disciplines and teams
- Fully Covered Medical, Dental, and Vision Insurance
- Paid Time Off
- Parental Leave
