Senior ML Infra Engineer for Distributed GPU Training

Genesis Molecular AI; City of Utica (NY); On-site; USD 150,000 - 190,000; Full time

Benefits offered by this job

Competitive compensation with salary +

Job summary

Genesis Molecular AI is seeking an experienced ML infrastructure engineer to lead the engineering efforts on our AI platform, focusing on scalable distributed infrastructure for training, inference, and evaluation across multiple clusters and clouds.

You will optimize GPU performance, accelerate workloads, and collaborate with researchers and scientists to advance our generative and predictive AI models for molecular space.

Qualifications

• Strong engineer with clean code and deep understanding of the codebases you work in.
• Deeply experienced with distributed training and inference of large models on GPU clusters and core libraries/frameworks: PyTorch, PyTorch Lightning, PyTorch Geometric, and Ray.
• Independent thinker with ownership and capability of engineering robust systems from first-principles to state-of-the-art realization.
• Curious, problem-oriented thinker excited by AI, physics, chemistry, and biology intersections with ML.

Responsibilities

• Lead engineering efforts focused on continuous improvement of the AI platform and scalable distributed infrastructure for ML training, inference, and evaluation.
• Support model training and deployment across multiple clusters and clouds, optimizing for throughput and cost.
• Optimize efficiency of ML models and workloads (latency, throughput, memory) including GPU performance tuning.
• Contribute to the long-term vision for Genesis’ infra platform.

Skills

Distributed training

GPU performance engineering

Ownership

PyTorch

PyTorch Lightning

PyTorch Geometric

Ray

Kubernetes

Terraform

CUDA

XLA

Triton

Tools

Kubernetes

Terraform

CUDA

XLA

Triton

Job description

Genesis Molecular AI is seeking an experienced ML infrastructure engineer to lead the engineering efforts on our AI platform, focusing on scalable distributed infrastructure for training, inference, and evaluation across multiple clusters and clouds.

You will optimize GPU performance, accelerate workloads, and collaborate with researchers and scientists to advance our generative and predictive AI models for molecular space.