Chimaraoke Mbata@ChimaM

RepLog was developed to solve the "data overwhelm" challenge faced by elite athletes and creators. While most logging apps focus on storage, RepLog focuses on transformation. By integrating the ibm/granite-3-8b-instruct foundation model, RepLog acts as a 24/7 AI coach "Bob" who analyzes raw training notes, physiological metrics, and sensory feedback to identify patterns that humans often miss. The application features a bespoke Glassmorphis UI designed to foster a state of "flow" during logging. Key technical innovations include a Multi-Session Architecture for parallel goal tracking, a Hierarchy-Aware Search Engine with real-time highlighting, and a User-Scoped Persistence Layer that ensures data privacy and account isolation while remaining entirely serverless on the client side. Whether you are tracking a 5km interval split or a coding sprint, RepLog provides structured feedback (Small Wins, Pattern Obs, and Next Focus) to ensure every subsequent "rep" is better than the last.

Build solutions that improve how software is built. Work with an AI that understands your codebase, helping you reduce repetitive work and build with real context. ⏳ Build your project in 48 hours. 🚀 Get hands-on with IBM Bob and explore how AI fits into real development workflows before it becomes standard. 💡 Build tools and workflows that developers would actually use. ⭐ Learn directly from experts throughout the event. 🤝 Join a team or build solo. 💰 Prize pool: $10,000 🧑🏻‍💻 Register before the Kick-Off Stream and start building with Bob.