Run more experiments faster ⚡️
Optumi is a cloud service that makes it easy for data scientists to develop & train ML models on powerful computing resources
Problem Optumi solves
Data scientists are often forced to wait a long time for ML training to complete and compromise on experimentation. This is typically because they run into resource bottlenecks on their laptops or cloud machines (ex: not enough RAM to try their full dataset). Optumi helps them easily scale to powerful instances with up to:
How it works
- Optumi is a JupyterLab extension that runs local Python notebooks in the cloud. We offer access to a wide variety of powerful cloud resources to launch interactive sessions or batch jobs.
1️⃣ Select a type of cloud machine
- Pick whichever machines suit your needs. We provision them and set up the environment to work like your laptop.
2️⃣ Select a mode of notebook interaction
- Connect to “sessions” for an interactive experience or launch “jobs” so notebooks keep executing when you close your browser.
3️⃣ See how your notebooks perform
- Get notified when notebooks complete or fail. We collect your output files and clean up cloud resources so you don't overspend.
Your benefits
Faster execution ✅
Reduce processing time by leveraging powerful CPU and GPU machines.
New experiments ✅
Leverage larger datasets and search through more hyperparameters.
No runtime limit ✅
Run notebooks for hours or days without worry. We don't kill them off.
Background execution ✅
Jobs enable your notebooks to continue running when you close your laptop.
Auto-release ✅
We clean up computing resources when jobs end to save you money.
Notifications ✅
Get text messages when jobs complete or fail so you don't have to babysit them.