The SQE replaced the LPC in 2021 with a promise: a single national assessment, fairer access to the profession, more candidates, lower cost. The qualification has delivered on the access part. The study materials around it have not kept up.
Most candidates we spoke to were stitching together five or six tools — a textbook for the syllabus, a question bank for practice, a flashcard app for retention, a chatbot for confused moments, a separate timer for mock conditions, and somewhere, a spreadsheet to track progress. None of those tools knew what the others knew. The chatbot couldn't see the questions you got wrong. The question bank couldn't pull from your tutor's notes. The flashcards weren't generated from your weakest topics.
SQEase started as a personal project in mid-2025. Finlay was supporting his girlfriend through her SQE1 preparation and kept running into the same problem: the existing tools weren't built for how real candidates actually study. He built a small pipeline that could turn her notes into practice questions that actually felt like the exam. The pipeline worked. The grades improved. Other people asked to use it.
We've kept building, but the original constraint hasn't changed: every question must come from material the candidate has actually studied, and every answer must be verifiable. No question is shipped that hasn't passed a five-stage critique. No explanation from Lex is shipped without a citation. If we can't tell you where a fact came from, we don't put it in front of you.
This is unfashionable in 2026. Most AI products celebrate fluency. We celebrate refusal — every dropped question is a question we've protected you from. That's the trade-off we're making, and it's the trade-off we think a regulated profession deserves.