Purvi Singh@purvisingh

Problem & Solution Statement: Joining an unfamiliar software project can be overwhelming for new developers. Before contributing code, a developer often has to understand the repository structure, identify important files, configure the environment, learn how different components interact, and determine which task is safe to work on first. This onboarding process can take hours or days, especially when documentation is incomplete or difficult to navigate. CodeSensei is an AI-powered developer onboarding copilot designed to make this process faster and more structured. It uses IBM Bob 2.0 Agent Mode to divide repository analysis into four specialized agents: Architecture Mapper, Setup Agent, Codebase Q&A Agent, and First-Task Finder. The Architecture Mapper explains the repository structure, important files, dependencies, and overall architecture in plain language. The Setup Agent analyzes the project requirements and provides a practical setup checklist and troubleshooting guidance. The Codebase Q&A Agent allows developers to ask questions about the actual repository and receive codebase-specific answers with file references. The First-Task Finder identifies practical beginner-friendly tasks and provides a primary task along with alternative tasks. The results produced by these agents are structured as JSON outputs and displayed through a simple dashboard containing Overview, Architecture, Setup, Ask, and First Task sections. The target users are developers, students, interns, and new contributors joining an unfamiliar codebase. Instead of replacing the developer, CodeSensei gives them the context and direction needed to become productive faster. What makes CodeSensei unique is its multi-agent onboarding workflow: different specialized agents analyze different onboarding problems in parallel, and their results are combined into one focused developer experience. CodeSensei aims to transform the first day of development from confusion and repository exploration into a clear path toward understanding and contribution. From “Lost on Day 1” to “Shipping in Days.”

💻 Online Hackathon | ⏳ Build: 48 hours | 📅 September 25–27, 2026 The official IBM Bob 2.0 hackathon. Build what's next in AI-assisted development, alongside the developers and experts shaping it. ⭐ Get help from experts throughout the event 🤝 Join a team or build solo 💰 Prize pool: $12,000 🧑🏻‍💻 Register before the Kick-Off Stream to start building with Bob

🌎 Hybrid Hackathon | 💻 Online Build: September 10–16, 2026 | 📍 On-site: September 15–17, 2026 The official hackathon of AI Infra Summit — co-organized by Kisaco Research and lablab.ai. Build production-grade AI applications on the modern AI infrastructure stack, alongside 8,000+ attendees, researchers, founders, and infra leaders. 📍 On-site Venue (September 15–17): Santa Clara Convention Center 5001 Great America Pkwy, Santa Clara, CA 📍 On-site participation is by invitation only. Travel and accommodation expenses will not be covered.