QueryForge Dashboards and AI Feedback
TL;DR
About six weeks later, QueryForge gained admin and student dashboards plus AI help so students could get feedback even when I wasn't available.
Benjamin Hyde
Education Leader & AI Builder
About six weeks after the initial QueryForge ship, the next major update added the student dashboard, the admin dashboard, and an AI help layer for instant feedback. This was the point where the project started to feel less like a SQL sandbox and more like a teaching tool.
The main reason for adding AI help was practical. I wanted students to be able to get feedback on their work when I was not available, especially when they were completing work at home or outside class time. Instead of being stuck until the next lesson, they could get immediate guidance and keep moving.
What changed
The student dashboard gave learners a clearer sense of progress and made the platform feel more personal. The admin dashboard gave me visibility into how students were working and where they were getting stuck.
The AI help feature added a support layer on top of the challenges. The idea was not to replace teacher feedback, but to make sure students could still get timely guidance when they were working independently.
Why it mattered
One of the biggest frustrations with student coding work is that momentum disappears when help is not immediately available. That is even more true when students are working from home. This update was about reducing that delay.
By combining dashboards with AI feedback, QueryForge became more useful both for students needing support and for me as a teacher trying to see progress across the class.
Screenshots
Build Notes
Approach
Build the next layer around visibility and feedback: give students a dashboard, give the teacher oversight, and add AI support for faster response when students are working independently.
Tools Used
Next.js, TypeScript, Prisma, SQLite, OpenAI
What Worked
The platform became much more useful once students could see their own progress and get immediate feedback without waiting for the next lesson.
What Failed
AI feedback still needed careful framing so it supported learning without encouraging students to rely on it as a shortcut.
What's Next
Keep refining the quality of the feedback, improve teacher visibility, and make the challenge flow feel even more seamless.