​
What are Evaluators?

With Evaluators, you can measure the quality of materials created by AI-powered edtech apps. Each set of evaluators:
  • Measures aspects of pedagogical alignment
  • Helps identify areas for improvement
  • Helps design consistent user experiences

​
Evaluators and Learning Commons

Evaluators is part of the Learning Commons’ AI Developer Tools for Education. Our goal is to help developers build more effective, evidence-based tools that support student learning and whole child development. We’re working to close the gap between cutting-edge AI and learning science by: Anchoring in learning science
We collaborate with experts at the forefront of learning science and rubric development, like Student Achievement Partners, CAST, and ANET. Working with pedagogical experts
We do the hard work of understanding how experts think and co-creating ground-truth datasets that reflect that expertise. Building, verifying, and shipping
We are building software that can automatically score any given text the way an expert would.

​
Our approach to building evaluators

We are building evaluators that rigorously measure a variety of dimensions that define what makes ‘good’ AI-generated material. We’ve built a rigorous, repeatable process for defining and building evaluators that edtech companies can use to inspect their AI-generated output.
Our Approach To Building An Evaluator Process Sv