Bar Buddy   Apply for the pilot

Tutor pilot · February 2027 cohort

You see them two hours a week. Bar Buddy sees the other sixty.

Your students study almost entirely without you, and you find out how it went by asking them. The tutor console turns those sixty hours into a ranked session brief — what leaked, why, and what to open with — and lets you push the work back into the app they already study in.

Apply for the pilot See the console

Sign in with Google to apply · Every tutor approved by hand · Students opt in and can revoke any time

Rule Sim — try it right here Live demo
TOR-BAT-01 Torts · Intentional torts

State the rule for battery.

Type it from memory. No hints, no multiple choice — that's the whole point.

0 words

For tutors

You get two hours a week. The console gives you the other sixty.

Most of a tutoring session is spent finding out what happened since the last one. Your student tells you they "did some Evidence" and felt okay about it. Bar Buddy already knows exactly what they did, what they got wrong, and — because of the confidence flags — which of their right answers were luck. You walk in already knowing where to start.

Pick a student — this is the brief you'd open with

Roster

Sam W. — invited 2 days ago, hasn't accepted. No data until they do.

Alex M. — ended. Passed July 2026. Access closed automatically on Aug 11.

Example data, fictional students. The brief is generated, not written — it ranks by tested weight × leakage, so a weak topic that carries more of the exam outranks a weaker one that barely appears.

Push work, don't assign it twice

Send a timed MCQ set, an essay rewrite with one element flagged as required, a rule-sim block, a flashcard deck, or a section of your own outline — straight into their Buddy.

  • You see pushed, seen, started, finished or skipped — so "I ran out of time" is a conversation, not a guess
  • Two skips in a fortnight raises a pattern flag before it becomes a month
  • The work lands where they already study, not in an email they archive

Your materials stay yours

Upload your own rule packs — your phrasing, your elements, your mnemonics — and your attack outlines. Grant them per student.

  • Your rule text is what they're graded against in Rule Sim, not ours
  • Outlines open in a view-only reader: no download, no print, no copy
  • Access expires with the engagement, automatically
  • Nothing you upload is used to train anything or shown to another tutor

Rule packs and outlines open to pilot tutors next

Built to survive a reference check

Every tutor is approved by hand before a single student is attached.

  • Every engagement runs on scoped, expiring, student-granted consent
  • Institutional conflicts are disqualified, not disclosed after the fact
  • The disclosure language is written for you and inserted automatically
What you see, once they grant it
  • Accuracy by subject and topic, with how much of the exam each carries
  • Confidence flags — including the right answers they weren't sure about
  • Essay element coverage and which elements they keep dropping
  • Activity: when they last studied, streaks, and assignments skipped
What you don't — unless they say so
  • Their conversations with the Buddy. Off by default, student-controlled only
  • If they do turn it on, transcripts are read live, never stored on your side, and every read is written to their access log
  • Anything after the consent window closes — the console goes dark on its own
  • Anything at all before they accept the invitation

Two ways to work with us

Resell it, or refer it. Pick once a year.

Both tracks run the same console and the same consent rules. The difference is only who bills the student.

Reseller

Buy seats up front at a volume discount, bundle them into your own package, and charge whatever your package is worth.

  • Three volume tiers; the discount deepens substantially from a handful of seats to a full cohort
  • You are the merchant. You invoice the student; we never contact them about price
  • Seats are good for one exam cycle, expire twelve months unredeemed, and are refundable for thirty days
  • No disclosure copy required — you're selling your own product

Fits a tutor who already sells packages

Affiliate

Hand out your code. The student buys from us at retail, and you earn a share of every payment for as long as they're attached to you.

  • Revenue share on every payment — cycle or monthly — not a one-time bounty
  • 90-day last-touch attribution, monthly net-30 payouts, low payout minimum
  • No inventory, no invoicing, no money out of pocket
  • FTC disclosure required wherever you recommend it — we write it and insert it for you

Fits a tutor who'd rather not handle billing

Your track is locked for a program year and switchable once, at a cycle boundary — and a switch never changes the terms on students already attributed to you. Seat prices and the retail price both go live when the doors open in November.

The reports

Your question bank knows you clicked C. It has no idea why.

Every other tool's unit of data is a letter. Ours is a letter, a confidence flag, and a sentence you wrote about your own reasoning while it was still live. That gap isn't a modeling advantage we might lose — it's data nobody else collects, and you can't reconstruct it after the fact.

Pick the flag you'd have used

What a question bank stores

What Bar Buddy stores

    So the report says

    Insight 01

    Your true score, not your comfortable one

    Every right answer you flagged as a guess or an unsure is scored as a miss, and a miss outweighs a lucky right. The number that comes back is lower than your percentage — and it's the one that predicts an essay.

    Raw 79% · True 64%. The 15 points between them are all in Evidence.

    Insight 02

    Decay, told apart from difficulty

    Because every attempt carries its own exposure history, the report can separate a rule you never had from a rule you're losing. Those need opposite responses, and a percentage can't tell them apart.

    You had CRP-07 in June and missed it twice since. This is decay, not a gap.

    Insight 03

    The cause, not the topic

    Four wrong answers in one subject can have four different fixes. The engine reads your own words and classifies each one — a rule you never had, a distinction you couldn't split, a fact pattern you misread, an outline you leaned on.

    3 of 4 Evidence misses were the same last step, not the same rule.

    Insight 04

    What not to do

    The most expensive habit in bar prep is over-correcting — re-drilling questions you already reviewed because one miss looked alarming. Every report names what to leave alone and why.

    Don't re-drill Torts. One miss in a 84% section is noise, and it cost a day.

    This is the part no one can copy by shipping a better model. A question bank can only ever know what you clicked, so it can only ever tell you a percentage. We ask one more question — how sure were you? — and read one more answer: why you think you missed it. Every insight above falls out of those two, and neither can be added to a dataset that never captured them.

    Tutor pilot

    Bring your students into the pilot.

    The console is open to a small group of tutors for the February 2027 cohort. Sign in with Google to apply — each tutor is approved by hand — then share your join link and your students connect in a minute.

    Apply with Google

    Rather read the partner brief first? Leave your email and we'll send it.