New in November 2021
Welcome to Baseten's very first monthly update. We’ll share these updates here on our blog and through our email newsletter—join our waitlist to get the latest product improvements, blog posts, and company updates delivered directly to your inbox.
Since we introduced Baseten, we’ve been busy working with users in our closed beta. We’ll share some of the cool things they’ve been building with Baseten very soon! In the meantime, keep reading for more Baseten updates.
Build with Baseten
Here are a few ways to start using Baseten today:
Deploy a fast.ai model: Deploy a simple fast.ai model with Baseten, ending up with a REST API to call the model. (Or, follow similar steps to deploy another type of ML model!)
Build a data labeler: Learn how to quickly build a user-facing data-labeling app, before you’re ready to deploy your model with Baseten.
Use a SageMaker model in BaseTen: Use Baseten to build a user-facing application with your existing SageMaker model, instead of deploying the model through Baseten first.
While we’re still in closed beta, we’re eager to get more of you using Baseten for free. If you don’t have access to Baseten yet, you can jump to the top of our waitlist by telling us a little about how you want to use Baseten here.
See what’s new on Baseten
We’re making changes daily, and we share new features and product improvements on our changelog. Here are a few things we’ve shipped in the last month:
User management: Invite others to see and edit your models and apps.
Protect your work: Prevent yourself from accidentally editing over your work (or a colleague’s work) when editing a worklet, query, or view in multiple windows.
Long-running tasks: Successfully run worklets with tasks that take hours by executing asynchronously.
Error messages: More descriptive error messages for worklet and node outputs.
Select component: Build a dropdown selector into your app.
Join our team
Our small team is hiring! We’re building towards a future where many more ML models get used to drive real-world outcomes instead of being abandoned in Jupyter notebooks. If you’re energized by what we’re doing and want to help us build something from the ground up, come join us.
