University lecturer builds RAG search over his school's regulations on top of a MAX bot

An IT department lecturer who has been writing code for more than twenty years described on Habr how he built a bot in the MAX messenger for his own needs. Over time it grew into a smart search system (RAG) over the university's regulations.

How the project started

The idea came from everyday irritation: the author was tired of opening the university website every time he wanted to check his own schedule. The solution was a bot in MAX that answers the simple question "what do I have today?" with the current class schedule.

• The author is an IT department lecturer who has been writing code for more than 20 years
• The bot runs in the MAX messenger
• Its original function was to answer "what do I have today?" with the schedule
• The project team grew to three people
• Its final function is smart search (RAG) over the university's regulations, a document hundreds of pages long

How a timetable became RAG search

The bot quickly caught on with users and acquired a team of three. With popularity came questions that the schedule simply could not answer: how to transfer to another program, how to be reinstated after expulsion, what happens with a scholarship, and what to know about dormitories. All of these topics are governed by the university's regulations, a document hundreds of pages long that is inconvenient to search by hand.

This need led the team to build smart document search, known in professional terminology as RAG (retrieval-augmented generation): a technique in which the system first finds relevant fragments of a document and then generates an answer based on them. In his article the author explicitly calls what he built "smart document search, a.k.a. RAG".

Why university regulations are a hard corpus to search

A university rulebook hundreds of pages long does not lend itself well to ordinary keyword search. Students phrase questions in everyday language ("what happens if I fail the exam session?"), while the answer may be written in legal language in a separate section of the document. This is a typical task for the RAG approach, where the system looks for relevant fragments of the document and builds an answer from them, rather than just returning word matches.

What this means

The story shows a practical path to adopting RAG in a small team. A technology usually discussed as a corporate AI tool grew here from a personal timetable bot into a working service that answers legally significant questions from university students.

Frequently asked questions

What is RAG in the context of this bot?

RAG (retrieval-augmented generation) is smart document search in which the system finds relevant text fragments and generates an answer from them. This is how the author describes the final version of his bot.

How did the bot project begin?

With a simple function: answering the question "what do I have today?" with the current class schedule in the MAX messenger.

Who developed the bot?

An IT department lecturer with more than twenty years of development experience. Over time a team of three people joined the project.