Innov8 4.0: A Hackathon by Eightfold Ai X Aries IIT Delhi

Welcome to Innov8! Hosted by ARIES, IIT Delhi in collaboration with Eightfold.ai, Innov8 is the ultimate convergence of creativity and technology. This event challenges you to leverage your innovation and technical skills to tackle real-world problems using machine learning models. Whether you're enhancing existing solutions or creating groundbreaking ones, Innov8 is your chance to showcase your ability to think outside the box and drive impactful change with AI. Ready to innovate? Rewards & Benefits: Possible Internship Opportunities: Top performers may secure coveted internships at Eightfold.ai, providing a gateway to turn your Innov8 success into a career breakthrough. Massive Prize Pool: Compete for a share of a ₹2.25L+ prize pool and earn exclusive merchandise as a reward for your creativity and technical brilliance. Compete with the Best: Test your skills against top minds, push your limits, and elevate your expertise. Eligibility: Open to students currently enrolled in undergraduate programs at IIT Madras, IIT Delhi, IIT Bombay, IIT Kanpur, IIT Kharagpur, IIT Roorkee, and IIT Guwahati, NIT Tiruchirappalli (Trichy), NIT Rourkela, NIT Surathkal, BITS(Pilani, Goa and Hyderabad), IISC Banglore, IIIT Delhi, NSUT, IIT Varanasi (BHU), IIT Hyderabad, IIIT Hyderabad, DTU, IISER(Mohali, Pune), TIET Patiala, PEC Chandigarh, NIT Kurukshetra, MNIT Japiur, IGDTUW, KIIT Bhubhneshwar, UPES, NIET, IIT Patna. Teams can consist of 2 to 5 members. All participants must register on the Unstop platform before the registration deadline. The problem statement and rulebook for the prelims have been released here on Unstop; please scroll down to the attachments section to access them. Final Submission Guidelines: Submit a single ZIP file containing the following: Transcript.txt Contains transcripts of all sessions Use original audio file names as identifiers (example: file.mp3 : text) Transcripts must only contain lowercase alphabetical letters and space (no punctuation, numbers, or special characters) Submission.json A single JSON file with complete analysis for all subjects combined Must strictly follow the prescribed format (see template link below) Source Code (.py) All Python scripts used in your solution If models were trained or fine-tuned, include only the training scripts (not weights or datasets) Readme Clear instructions to run the code Mention Python version, virtual environment usage, required packages (pytorch, transformers, etc.), and installation commands Note: Submissions missing either transcript.txt or submission.json will not be evaluated Evaluation will be done only on the official evaluation set