Founder work / Structured problem solving with rubric scoring
BitMento ProblemLab
A structured problem-solving experience where learners work through generated problems and get their solutions scored against a clear rubric.
- Status
- Built as part of the BitMento ecosystem
- Year
- 2026
A representative design of the idea, not a screenshot of a finished product.
My role
- Designed the assessment workflow and the rubric-based scoring approach.
- Built the interface and the access-code system used to give people entry.
Practice problems are only useful if the feedback is. Vague feedback leaves learners guessing what to improve.
Why it mattered
A visible rubric tells a learner exactly where a solution was strong and where it fell short, which makes the next attempt better.
- Generate the problem, collect the solution, score it against defined criteria, explain the score.
- Gate access with codes so a pilot group can be managed.
What it does
- AI-generated problem statements.
- Rubric-based evaluation of submitted solutions.
- Code-based access for controlled pilot groups.
What I wanted to build: Practice problems with feedback that explains itself.
How it came together
- Defined the rubric, wrote prompts that generate problems and score solutions against it, then tuned for consistency.
What I did
- Defined the requirements and the user journey before any code was generated.
- Made the design and product decisions, and reviewed what the tools produced.
- Tested, debugged and decided what to keep, rewrite or cut.
- Configured the services and handled deployment.
Where AI tools helped
- AI tools drafted and refactored code from structured prompts, and I iterated on the results.
Services connected
- Supabase for access codes and data, Groq and Gemini models for generation and scoring.
Working today
- Problem generation, rubric scoring and code-based access.
Still prototype or planned
- Scoring consistency is still being refined.
Written from my own account of the build.
- Making AI scoring consistent enough to feel fair.
A working assessment workflow inside the BitMento ecosystem. No usage numbers are claimed.
Transparent criteria build more trust than a clever score.
Need an assessment or evaluation flow built?