Data Genesis 2026

Build the Data. Build the Intelligence.

Evaluation only · Open

Data Genesis 2026 is a national-level AI dataset creation and computer vision hackathon organized by the Department of Computer Science and Engineering, Ramco Institute of Technology, in association with the RIT IEEE CS Student Chapter and RIT GFG Student Chapter. The hackathon challenges participants to create high-quality, India-centric datasets from scratch and transform them into intelligent computer vision models.

Overview

Unlike conventional hackathons that provide ready-made datasets, Data Genesis 2026 places data creation, curation, and AI development at the centre of the challenge. The competition consists of two rounds: Round 1 — Dataset Development Teams select an official dataset theme, capture original images, annotate and organize them, prepare documentation, publish the dataset on Kaggle, and submit the required links and documentation. Round 2 — 24-Hour AI Hackathon Qualified teams attend the in-person hackathon at Ramco Institute of Technology and develop an object detection and recognition model from scratch using their own Round 1 dataset. Participants cannot use external or benchmark datasets, pre-trained models or weights, transfer learning, foundation models, or existing trained object detection models. Models must be developed using the team's own dataset with randomly initialized weights. The hackathon will be conducted in two rounds: Round 1 (Online Dataset Submission), with a deadline of September 14, 2026 and results announced on September 19, 2026. Teams must create and publish an India-focused dataset reflecting culture, heritage, agriculture, or indigenous knowledge on Kaggle to qualify. Shortlisted teams will advance to Round 2 (24-Hour In-Person Hackathon), taking place on September 24–25, 2026, at the Ramco Institute of Technology campus, where they will build an object detection and recognition model using strictly their own created data.

The brief

Create an original, machine-learning-ready dataset based on one of the official themes and use that dataset to develop an object detection and recognition model from scratch.

Dataset Themes

Indian Culture & Heritage

Confectionery of India Traditional Indian Musical Instruments Indian Handloom Textures & Patterns Indian Leaf Plates Indian Toys and Games Traditional Indian Footwear Traditional Indian Lamps Indian Temple Bells and Ritual Objects Indian Terracotta and Clay Artifacts Traditional Indian Baskets and Woven Crafts Indian Pottery and Earthenware Traditional Indian Jewellery Designs Traditional Indian Self-Defence Implements Indian Attire Through the Ages Traditional Indian Games & Entertainment

Indian Agriculture & Natural Resources

Fruits & Vegetables from Indian Markets Indian Millets, Pulses, and Grains Indian Seeds and Seed Varieties Indian Medicinal Leaves and Herbs

Specialized Computer Vision Dataset

Wrist Veins

Deliverables

Round 1 - Dataset development plan, Original image dataset ,Image annotations, Dataset organization,Dataset documentation, Published Kaggle dataset, Kaggle link and required submission documentation
Round 2 - Preprocessed dataset, Object detection and recognition model, Model training pipeline, Model evaluation, Inference demonstration, Technical presentation

Evaluation criteria

Round 1 — Dataset Development
Dataset Planning — 15%, Image Quality & Diversity — 40%, Annotation Quality — 20%, Documentation & Organization — 20%, Kaggle Publication — 5%
Total — 100%
Round 2 — AI Model Development; Model Architecture & Innovation — 30%, Training Pipeline — 15%, Performance on Hidden Test Data — 45%, Presentation & Demonstration — 10%
Total — 100%

Files and datasets

DOCUM
Data_Genesis_2026_Key_Information
It is a one-page summary of DataGenesis 2026 covering the hackathon format, eligibility, deadlines, themes, rules, evaluation, prizes, fees, and contact details.
39 KB
Download

Frequently asked questions

Who can participate?
Students, researchers, professionals, and AI enthusiasts can participate.
Is there any age restriction?
No. There is no age restriction.
What is the maximum team size?
A maximum of 3 members can participate in one team.
What happens in Round 1?
Teams select a theme, create an original dataset, capture and annotate images, document the dataset, publish it on Kaggle, and submit the required details.
What happens in Round 2?
Qualified teams participate in a 24-hour in-person AI hackathon at Ramco Institute of Technology, where they develop an object detection and recognition model from scratch.
Can teams use pre-trained models or external datasets?
No. External or benchmark datasets, pre-trained models or weights, transfer learning, foundation models, and existing trained object detection models cannot be used.

More challenges at Department of Computer Science and Engineering, Ramco Institute of Technology

Prizes are offered and awarded by Department of Computer Science and Engineering, Ramco Institute of Technology, not by Wooble. Amounts shown are as stated by the host and may include the host’s own valuation of non-cash items. Who places, how and when a prize is paid, and any tax or deduction, are between the winner and the host. A challenge may close, change or be withdrawn before results are declared.