Research, field notes, and technical depth from the team building AI evaluation for Indian education.
Showing 1-9 of 30 posts
Grading automation gives you a score. Evaluation infrastructure gives you a memory layer. The distinction determines everything you can build downstream.
EdTech generates its richest data at the point of evaluation — then discards it. What lives inside a student's answer, and why capturing it changes everything.
Field study: AI evaluation of open-ended Hindi handwritten answers achieved α=0.808 reliability across 11 tests in MP and Rajasthan. What the feedback quality data actually shows.
A research-led analysis of mismatch patterns in handwritten subjective evaluation—and what they reveal about fairness, governance, and scale.
A research-led perspective on how governed, teacher-led AI evaluation systems can deliver transparency, trust, and scale in public education.
Design patterns for appeals, human-in-the-loop moderation, and learning-oriented dispute resolution that protect fairness and scale.
Engineering and data-ops patterns for throughput, latency, offline scoring, and multi-tenant deployments in public systems.
A playbook for generating pedagogically useful AI feedback—specific, scaffolded, rubric-aligned—and measuring teacher trust and adoption.
Step-by-step benchmarking playbook: sampling strategies, inter-rater agreement metrics, and how to interpret gaps between AI and human scoring.
We use cookies to enhance your browsing experience, serve personalized content, and analyze our traffic. By clicking "Accept All", you consent to our use of cookies. You can manage your preferences or learn more in our Privacy Policy.
'/%3E%3C/svg%3E)