Openlayer is a powerful testing and observability platform for ML. It lets you collaborate with others on finding issues in models and data, debugging them, and committing new versions.
Very cool approach to ML testing. I like how you track against commits and help define goals as you define the pipeline. One question - how do define the "root cause" that you mention when solving failed goals?
This is awesome! Having a great debugging workspace on par with software engineering debugging has always been a pain point to me when working on finance data and autonomous driving. What are some of the use cases you enable today?
Hey @lawlm thanks! Companies and orgs in the finance space stand to benefit greatly from Openlayer. Many of the people we work with use us to help de-bias and improve the accuracy of models that predict whether to give loans, or whether a transaction is fraudulent, for example. These use cases are especially important because they have a large impact on people's lives, so it's critical to invest in evaluating these models.
Great integrated product for giving visibility into ML models, highly recommend for anyone looking for a way to benchmark, evaluate, and iterate on their models (which everyone should!)
Openlayer