Built by students, for the people who look after students
Ascora was built for KPMG in India's Smart Campus Analytics challenge at HacXLerate 2026 by byteXL.
Why we built it
We have seen classmates who looked fine on a marks sheet drift away quietly: fewer logins, missed labs, skipped mock interviews. By the time a grade drops, the moment to help has often passed.
Colleges already hold the signals, spread across seven different systems. Ascora brings them together, explains who needs support and why, and gives mentors a realistic plan for the week.
The team
Five builders, one semester in
All five of us are B.Tech AIML students in our second year (Semester 3) at CU UP campus.
Taqi
Team lead
Collaboration, coordination and decision making: set the direction, kept the team in sync and made the calls.
B.Tech AIML, 2nd year, Sem 3
Ashmit Tiwari
Branding, aesthetics and style
Brand identity, logo teaser, and the look, feel and visual style of the platform.
B.Tech AIML, 2nd year, Sem 3
Harsh Gupta
Branding, aesthetics and style
Promo graphics, theme and UI styling, and the visual story judges and colleges see.
B.Tech AIML, 2nd year, Sem 3
Lakshya Singh Parihar
Synthetic cohort and database
The synthetic campus cohort and the database it lives in.
B.Tech AIML, 2nd year, Sem 3
Ujjwal Gupta
Backend APIs and ML models
Python and FastAPI routes, and the machine learning models behind the scores and risks.
B.Tech AIML, 2nd year, Sem 3
What we believe
Three rules we built Ascora around
Action over alarms
A flag is only useful if it leads to a conversation, a plan and a check on whether it worked.
Explain everything
Every number in Ascora can be traced back to its inputs, in words a mentor can repeat to a student.
Students first
Scores are for the people who support students, never labels that follow a student around.
About the data
Every student in Ascora is generated for this project. The cohort of 1,200 students across seven campus systems includes realistic problems on purpose: inconsistent roll numbers, typos, duplicates and conflicting records. No real student data is used anywhere.
From early warning to real intervention
Sign in as a head of department or a faculty mentor and walk a full cohort of 1,200 students.