Ten challenges built to separate “it works” from “it works brilliantly.” Four selected from SIH 2024 and six original CODE HEIST challenges.
Develop a system that automatically generates videos using frame interpolation techniques for WMS services, improving the temporal experience of geospatial visualisation.
Develop an AI/ML-based map-matching system capable of distinguishing vehicle movement on a highway from movement on an adjacent service road despite imperfect location data.
Build an innovative digital-forensics and incident-response tool that helps investigators rapidly collect, organize and analyze digital evidence.
Enhance extremely low-light imagery of permanently shadowed lunar regions captured by Chandrayaan-2's Orbiter High Resolution Camera.
People use WhatsApp, emails, PDFs, screenshots, notes and calendars to store information, but important information gets lost across different platforms. Build an AI assistant that creates a personal searchable memory from user-provided data.
Build a digital twin of a university campus representing buildings, classrooms, labs, hostels, parking, events and facilities in one interactive system.
Build a real-time payment risk assistant that evaluates transaction context and explains why a payment may be unusual or suspicious.
Build an AI-powered energy intelligence platform that identifies unusual consumption, energy wastage, peak demand and abnormal usage across buildings, rooms or devices.
Medical information can be scattered across prescriptions, laboratory reports, scans and hospital visits. Build a system that converts multiple medical documents into a longitudinal patient timeline.
Build a platform where a user gives a high-level objective and the system decomposes it into multiple tasks and executes them using available APIs and services.
Some challenges can produce a working MVP quickly. The real difficulty is pushing beyond the obvious solution — improving accuracy, reliability, scalability, explainability and real-world usefulness.