sorry can't get this for you

2 minutes read

February 24, 2023

Pipeline AI product update: February 2023

Explore our 10% user growth, enhanced API scaling, new custom environment features, and the latest on pipeline versioning.

Welcome to Feb's newsletter. We've had some good increase in number of users over the past month (+10% growth) and this means we've been busy making sure the API kept scaling without any issue.

Further to our work on improving API's reliability, we have focused on launching new features! Last month we launched pipeline versioning via our tags system, if you haven't used it yet, you can see how on our docs. This month we've been very focused on launching one of the most sought after features, custom environments. This means you can now upload and run your own pipelines with your own custom environment and libraries! - More on this feature below.

What's next? 

Our development roadmap for Q1 includes further API optimisations and feature additions in response to feedback and requests from users and include:

  • 🔜 API deployment to multiple clouds (Under construction)
  • 🔜 API V3... 

We will keep doing monthly product updates, but to keep up to date and for immediate support from our team, make sure to join our Discord server.

Call for conversations in the Finance sector!

We are about to launch a new product specifically for the finance sector. We will be at the AI in Finance Summit in New York - let us know if you are going to be there and we can book a meeting.

New feature: Custom environments

Until now, whenever you uploaded your pipeline to our cloud, runs performed on that pipeline used a default environment with a set list of python requirements.

As of today, you can create and build out your own custom environments, adding any packages and versions of packages you see fit.

Using our CLI, creating a new environment is as simple as:

(on your shell)

$ pipeline environments create my-environment

You can then add packages to your environment:

(on your shell)

$ pipeline environments update environment_ID add numpy==1.24.2 torch==1.13.1

and upload your pipeline e.g. "my-pipeline" to our cloud, specifying that it should use that environment:

(on your python script)

api.upload_pipeline(my-pipeline, environment="environment_ID")

You are free to change any packages in that environment at any time. Other features such as locking your environment are also supported. We recommend you start exploring this new approach for a better experience.

More about this in our docs.

Newsflash: we are opening a London office!

Our team is growing and we will shortly be opening a London office. We are keen to get involved in the local scene and meet our users so feel free to invite us to meetups. We will happily sponsor/buy the beers! :) 

That’s it for this month. Thanks for reading and we’ll be back with more features around next month. If you'd like to give us feedback on anything we've shipped this month, just get in touch - we'd love to hear from you!

 We will keep doing monthly product updates, but to keep up to date and for immediate support from our team, make sure to join our Discord server.

Oscar Rovira

Co-founder & CPO


Pipeline AI makes it easy to work with ML models and to deploy AI at scale. The self-serve platform provides a fast pay-as-you-go API to run pretrained or proprietory models in production. If you are looking to deploy a large product and would like to sign up as an Enterprise customer please get in touch.