Data Science Hackathon

About the Opportunity: sciFI (scifi.ink) is hosting a free, online, 36-hour Healthcare AI Hackathon open to ML enthusiasts, data scientists, and developers worldwide. Participants will build an AI/ML model to detect anomalies in patient vital signs (heart rate, blood pressure, SpO2, temperature, etc.) — a real-world problem that sits at the intersection of machine learning and clinical care. All participants get free GPU compute and an AI copilot inside sciFI's ML notebook — no local setup required. Guidelines: Eligibility: Open to all — students, working professionals, and independent developers are welcome Team Size: Solo participation allowed; teams of up to 3 members permitted Inter-college teams: Allowed Inter-specialization teams: Allowed (Engineers, med students, data analysts — all welcome) Problem Statements: 1 core problem statement — Anomaly Detection in Patient Vital Signs Format: Online, async-friendly — work on your own schedule within the 36-hour window Platform: All submissions run on sciFI's AI-native ML notebook (browser-based, no install needed) Dataset: Provided at the start of the hackathon Submission: Jupyter notebook + short write-up explaining your approach Rules: All work must be done during the hackathon window (April 3–6, 2025) Pre-trained models and open-source libraries are allowed Use of public datasets in addition to the provided dataset is permitted Plagiarism or reuse of prior work will result in disqualification Each participant/team can submit only one final solution Submissions must include working, reproducible code Participants must sign up at scifi.ink before the start date Duration: Total Duration: 36 hours Round 1: Build & Submit (36 hours) Evaluation & Results: Within 48 hours of the submission deadline