Software Engineer - ML Infrastructure

Genesis Molecular AI is pioneering foundation models for molecular AI to unlock a new era of drug design and development. They are seeking experienced ML infrastructure engineers to lead engineering efforts focused on driving forward their ML research agenda for generative modeling of molecular systems.

Responsibilities:

• Lead engineering efforts focused on continuous improvement of the AI platform, focused on rapid build out and iteration on scalable and robust distributed infrastructure for ML training, inference, and evaluation
• Support model training and deployment across multiple clusters and multiple clouds, optimizing for throughput and cost
• Optimizing efficiency of ML models and other workloads in terms of latency, throughput, memory consumption, etc. (e.g., via GPU performance engineering), pushing the limits of what’s possible with the current hardware
• Contribute to the long-term vision for Genesis’ infra platform

Requirements:

• Strong engineer who constantly strives for technical excellence
• Ability to write clean code and have a deep understanding of the codebases you work in
• Deeply experienced with distributed training and inference of large models on GPU clusters
• Experience with core libraries and frameworks such as Pytorch, Pytorch Lightning, Pytorch Geometric, and Ray
• Independent thinker with a strong sense of ownership
• Capability of engineering robust systems from first-principles-based conceptualization to state-of-the-art realization
• Curious, problem-oriented thinker excited to dive deep into the emerging field at the intersection of AI, physics, chemistry, and biology
• Experienced with building, maintaining and debugging low-level cluster infrastructure running on multiple clouds using Kubernetes and Terraform
• Experienced GPU engineer who can quickly figure out performance bottlenecks and architect highly performant code for large scale ML workloads
• Experience with XLA, Triton, CUDA, or similar accelerator programming languages and/or deep learning compiler stacks
• Experience working with molecular systems (protein sequences and 3D structures, small molecules, etc.), ML force fields or other physics-informed models and methods, or point cloud data in other application domains, such as 3D graphics