FREE — INVITATION ONLY
S$0
Personally invited by Lucas. Limited slots. Apply for an invitation through the pilot program.
- Haiku 4.5 model
- Drill & Quiz modes
- Citation panel
- VOSE ID portability
- Discord community access
Verdict is built for serious students who want to engineer their own learning. Transparent retrieval. Just-in-time teaching. Citation-grade auditability. Made for power users — we never dilute the learning process.
These aren't features you can bolt on. They're decisions that shape the entire product surface. Competitors structurally can't match them without rebuilding from the core.
Every answer carries the source chunks it was built on. You can read them. You can disagree with them. You can copy the citation into your essay. There is no "trust us, the model knows" — because the model doesn't, and pretending otherwise is the original sin of AI tutoring.
Most tutors front-load the content, then test you on it. Verdict presents the question first. You attempt. You stumble. The teaching arrives at the moment of cognitive readiness — when the brain is asking, not when the curriculum says.
Every grade is reproducible from the chunks. You can audit your own Verdict. You can audit ours. If the citation doesn't support the claim, the verdict doesn't stand. This is what citation-grade means — not a footer link, an actual chain of evidence the model has to honor.
// the difference between "AI tutor" and a tutor that uses AI.
The three differentiators are architectural. These are craft. Built because the founder uses Verdict daily to study for his own A-Levels.
Drill, Quiz, Chat, Past-Paper, Mock — one wheel, one keystroke, no context switching.
Proficiency tracking by sub-topic. Color-coded attempt history. Failure modes are signal, not red marks.
Analytical (default) · Executive · Conversational · Mentor · Drill Sergeant. Pick how you're spoken to.
Source chunks live in a side panel you can pin, copy from, dismiss. The chunks are first-class UI, not an afterthought.
Why do serious players prefer Lichess over Chess.com?
Lichess shows you the engine eval. Win probability after every move. The score of every line you didn't play. Chess.com hides the math behind a "premium" wall and a friendlier interface. Both are popular. Only one is preferred by serious players.
The same divide exists in AI tutoring, and most products are on the wrong side of it for serious learners.
Lichess didn't win by being faster than Chess.com. It won by trusting its users to want the truth. Verdict is built on the same bet.
I'm Lucas. I'm sitting H2 Economics in November 2026 as a private candidate — RJC alumnus, self-directed since.
Verdict started as my own study tool. The version I needed didn't exist, so I built it. It turned out other people needed it too. So I kept going, in public, in parallel with studying.
It's not the polished narrative of a 40-year-old EdTech founder with a PhD and a deck. It's a 19-year-old who wanted a better way to study and refused to wait for someone else to build it.
No "Enterprise" placeholder. No "Most Popular" badges. One Pro tier built for the people who actually use it; one selective Free tier for the people we ask to join personally.
Personally invited by Lucas. Limited slots. Apply for an invitation through the pilot program.
The full Verdict surface for the people who study with it daily. No feature gated for "enterprise reasons."
// no third tier. enterprise pricing is a wedge-2 conversation, not a wedge-1 promise.
Verdict's pilot is for the first students, parents, tutors, and educators who'll shape the product. Free tier is invitation-based; pilot Pro members get bonuses and direct founder access.
Command your learning.
Most products want you to relax. This one wants you to engage. Pick the side of the divide you want to be on.