Solomon
study trainer / mastery under pressure
Solomon is a high-rigor study trainer for dense subjects. It turns raw material into contextual cards, pressure-tests them with scenario prompts, and schedules review exactly when memory starts to decay.
5 stages
Import to mastery in one deliberate path.
1d-14d
A fixed review cadence for the first production cut.
0 leakage
Subjects stay isolated by route and data scope.
/app/biochemistry
ATP synthesis under variable oxygen conditions
/app/operating-systems
Deadlock prevention and scheduling tradeoffs
/app/calculus
Constraint optimization with Lagrange multipliers
The first production goal is simple: accept source material, extract concept candidates, let the user correct the output, and publish only the cards that survive approval.
Drop zone
Lecture PDF, notes, or text dump
Lecture 05.pdf
31 pages / enzymes
Kernel notes.txt
84 concepts found
limits-review.md
12 cards awaiting approval
Ship this path before groups, achievements, or social pressure mechanics.
PDF, notes, or pasted text arrive here before extraction and cleanup.
A short diagnostic removes material that the learner already owns.
Each concept stays attached to an expert example and visual anchor.
What-if prompts test transfer, not symbol recognition.
Cards return at 1d, 3d, 7d, and 14d until recall stabilizes.
What-if scenario
high transferPrompt: What breaks first if oxidative phosphorylation loses oxygen for 90 seconds?
Hint 1: map the bottleneck, not the symptom.
Hint 2: connect electron transport to ATP yield and lactate pressure.
Resolution: convert the answer into a reusable causal explanation.
Keep friction high while the trace is new.
Look for hesitation and weak transfer.
Require explanation, not just the answer.
Archive only if the concept survives scenario shifts.
This is the architectural rule that protects the learning model. Biology, operating systems, and calculus can live in the same account without contaminating each other's queues.