Most school software added a chatbot. We built the AI into the work.
A chat box that answers questions about a manual is not AI in a product — it is a search engine with a personality. Ilmora uses AI in five places where it changes what somebody actually has to do, and in every one of them the answer traces back to something you can open and check.
Ask Ilmora — the assistant
Nobody reads a manual. They ask.
A help button on every screen that answers in your context — what this screen does, where the action you want lives, and why the system just refused something. It takes you to the screen rather than describing the path, and hands you to a human when it should.
The report analyst
200+ reports, reachable in plain language.
Ask a question and Ilmora finds the report that answers it, runs it under your own permissions, and explains what the numbers show. It cannot invent a figure, because every answer is a real report you can open, export and check.
Agentic actions
Describe the outcome. Ilmora fills the forms.
Tell Ilmora what you want done and it performs the steps — generating challans for a section, scheduling a PTM, issuing admit cards. Every action is previewed before it runs and recorded against your name afterwards, because an agent that acts silently is an agent nobody can supervise.
Automated alerts and warnings
The system tells you, before somebody has to notice.
Ilmora watches the patterns that matter and raises them — a student sliding into chronic absence, a fee cohort falling behind, a register nobody marked, a hostel round that fell short. Delivered on the channel you chose, inside the hours you allowed.
Insights and forecasting
What is about to happen, not just what did.
Collection forecasting, enrolment and churn trends, staff and student performance movement, and campus comparison — with the working shown. Every figure traces back to the events that produced it, so an insight can be argued with rather than believed.
Three rules that make an AI safe to put in a school
It inherits your permissions
Not a filter applied after the fact — the assistant runs the same query behind the same gate. There is no privileged path for it to take.
Every answer is checkable
A figure always comes from a report you can open and export. The AI picks and explains; it never produces a number of its own.
Actions are previewed and recorded
An agent shows you what it is about to do before it does it, and signs what it did with your name. An agent nobody can supervise is not one you should run a school on.
About the AI specifically
Can the AI see data I am not allowed to see?
No. Ask Ilmora runs the same reports behind the same permission gates you have. If you could not open a staff disciplinary register yourself, the assistant cannot surface one to you either — it is not filtering an answer, it genuinely cannot run the query.
Does the AI make up numbers?
It cannot. Every figure comes from a real report you can open, export and check for yourself. The assistant chooses which report answers your question and explains the result; it does not generate the result.
What does 'agentic' mean, and is it safe?
It means you describe an outcome and Ilmora performs the steps — generating challans, scheduling a meeting, issuing admit cards. Every action is previewed before it runs and recorded against your name afterwards. It is in beta and starts with a small set of reversible actions.
Is our institute's data used to train AI models?
No. Your data is used to answer your questions and nothing else. It is not pooled with other institutes and it is not used as training material.
See the AI answer a question about your own school
Bring a question you cannot easily answer today. We will ask Ilmora during the demo, on data that looks like yours.