Supercharge Your Deep Learning Development Cycle

Customer Benefits

Up to 100x faster Deep Learning Training

Save 2 hours / day for each machine learning engineer

Up to 30% accuracy improvement from automatic model tuning

Fully Managed Infrastructure for Cloud and On-Prem

  • Easily scale from local GPUs to cloud GPUs.
  • Launch hundreds of deep learning jobs from command line without having to manage infrastructure.

Up to 100x faster Deep Learning Training

  • Mixed precision training.
  • Easy to setup distributed training across hundreds of GPUs.
  • Automatically fine tune your models using Neural Architecture Search (NAS).

Deep Learning Experiment Tracking and Insight Analysis

  • Track and evaluate thousands of models and quickly surface the best performing ones across your organization.
  • Automatic model reporting tools.

The Team

Founding Team

Jason Ge

PhD candidate, Princeton University

Davit Buniatyan

PhD candidate, Princeton University