Blood, Sweat & Tears
AI-marked training for the Lagree method
Blood, Sweat & Tears, a London Lagree studio and training brand, needed to scale how it teaches the Lagree method. CodeLeap built the training app behind it — a flashcard learn library, quizzes marked by AI, and a dashboard to track how learners progress.
What we did CodeLeap built the flashcard learn library, the AI grading that marks both spoken and typed answers using speech-to-text, an AI routine builder and the instructor dashboard, so assessment scales well past what one instructor can mark by hand.
Blood, Sweat & Tears runs Lagree classes and trains instructors in the method, a discipline with more than 550 distinct moves. Teaching that by hand and in person caps how many instructors the studio can produce. They partnered with CodeLeap to build a training app that teaches the moves, tests them with AI marking, and shows the studio how each learner is doing.
Product features
-
Learn move library
A digital flashcard library of the method's 550-plus moves, organised so instructors and students can study each exercise, its name, form and detail, and build recall of the full vocabulary at their own pace. The library is the reference the rest of the product is built on.
-
AI-graded quizzes
Quizzes test knowledge of muscles worked, form, transitions and common mistakes, with learners answering by typing or speaking. Spoken answers are transcribed and an AI marker accepts typos and wording variations, so a learner who knows the answer is not marked wrong for phrasing it differently.
-
AI routine builder
An AI tool drafts class routines from the move library, automating most of the routine writing. The studio can produce far more sessions each month while keeping the method consistent across all of them.
-
Admin and progress dashboard
An admin panel lets the studio manage quiz content and learning order, approve accounts and track engagement across screen time, sessions and scores, filtered by date. It turns instructor training from something felt to something visible.
Challenge
The hard part of this product is marking. A written quiz on 550 moves can be graded mechanically, but the knowledge being tested is spoken and practical, and a learner who says the right thing in their own words has answered correctly. Marking strictly would have made the quizzes frustrating and useless as a measure. The system needed to handle spoken answers, transcribe them reliably and mark them with enough tolerance to be fair, while still telling the studio something true about who has mastered the method.
Solution
CodeLeap built the flashcard learn library, the quizzes with AI marking of both spoken and typed answers, the AI routine builder and the admin and progress dashboard. Speech transcription feeds the AI marker, which accepts typos and variations in wording, so assessment scales well past what one instructor can mark by hand without becoming pedantic about phrasing. The studio now has software that teaches and tests the Lagree method and shows who has mastered it, instead of relying on in-person mentoring alone. It can train more instructors and build more routines without losing the consistency the method depends on.
Tech stack
The tech chosen was our speciality, enabling fast development of a reliable AI-assisted learning platform.
- Application
- Web app with student and admin dashboards
- AI
- Speech-to-text transcription with LLM-based marking, plus an AI routine builder
- Hosting
- Cloud infrastructure
Keep exploring
-
Simple PTPersonal trainers lose a large part of their week to scheduling admin and chasing payments, which is unpaid work displacing the coaching they earn from. CodeLeap designed and built Simple PT for iOS and Android in two months, on a tight budget.
read more -
A coach in every pocket. AI fitness & nutrition coachWe worked with one of the UK's leading protein brands to build an AI coaching app, centred on a conversational coach that adapts to each user's goals.
read more -
UdokkUdokk saw growing demand for trustworthy digital health content and had an extensive professional network ready to create it. CodeLeap built the platform where healthcare professionals publish scientific and educational video for practitioners, students and the public, and earn from it.
read more
