Principal GenAI Specialist Solutions Architect, Training & Inference

Communication Libraries, Frameworks, and Models

When it comes to communication libraries, we utilize NCCL and MPI. Our work also involves frameworks like PyTorch, NeMO, and Jax, along with models such as Llama and Nemotron.

We heavily engage with containers such as Docker and Enroot, as well as orchestrators/schedulers like EKS.

BASIC QUALIFICATIONS

• Bachelor's degree in computer science, engineering, mathematics, or related fields
• Deep familiarity across the stack is required, including:

• Compute infrastructure such as Amazon EC2 and Lustre
• ML frameworks PyTorch and JAX
• Orchestration layers like Kubernetes and Slurm
• Parallel computing with NCCL and MPI
• MLOPs
• Experience with target use cases in the cloud

You will be responsible for driving the development of...