AI4BI от Raft: ИИ-аналитик для тренеров и скаутов в футболе, хоккее и баскетболе

In 2026, Raft described the development path of AI4BI — an artificial intelligence platform for sports analytics, created in 2017 for cricket and grown over nine years into a universal tool for clubs across five sports and the mass fitness market.

How AI4BI Reached Its Current Scale

Raft built the first AI4BI prototype in 2017 specifically for cricket — a sport practically exotic in the Russian market. According to the team itself, the solution at the time looked like "some kind of spaceship": the market was not yet accustomed to AI analytics in sports.

Over nine years, the platform evolved from a single discipline to a comprehensive product:

• 2017 — launch of the first prototype, sport: cricket
• 2026 — support for football, hockey, volleyball, basketball, and bandy
• RAG-based knowledge bases for connecting domain expertise without retraining the model
• AI/ML module for forecasting, pattern analysis, and recommendations
• New branch — "civilian" fitness analytics for the non-professional market

For Which Roles Is AI4BI Designed

AI4BI is built not as a single universal dashboard, but as a role-based system: each user receives their own data layer and operational logic. A coach sees workloads and tactical patterns, a scout gets candidate comparison tools, a sports analyst gets deep cuts by matches. A sporting director and club owner work with a financial-strategic picture tied to results.

The key technical feature: the RAG mechanism allows uploading club- or league-specific information — tactical regulations, historical matches, medical protocols — and the model begins answering questions taking this context into account without retraining. According to Raft, this lowers the implementation threshold: domain expertise can be connected without the involvement of ML engineers.

"This is now an accessible tool for sports organizations from professional teams to fitness clubs," the

Raft corporate blog notes.

Three Cases — Three Sports, Three Roles

The central part of the AI4BI presentation consists of three demonstration use cases, each covering a separate sport and a separate user role. This combination shows how dramatically the interface and platform logic change depending on the context: analytics for a hockey scout is built around fundamentally different metrics than the toolkit for a volleyball coach.

The fitness branch of AI4BI operates on the same architectural foundation, but with different data: not game match statistics, but biometric indicators and individual client dynamics. A fitness club trainer receives recommendations for program adjustments, while a client gets a visualization of their own progress.

What This Means

AI4BI is a practical example of AI "domainization" through RAG: instead of a separately trained sports LLM — a universal AI engine enriched with subject-specific knowledge bases. This approach allows horizontal scaling — from cricket in 2017 to any sport and fitness in 2026 — without the cost of retraining for each discipline.