Shao Ern Ng@shaoern

Bug Squad is an IBM Bob-powered debugging pipeline designed to take a bug ticket from initial report to a reviewed and tested fix while answering an important question: what else could this change break? The system divides debugging into four specialized Bob subagents: Reproducer, Investigator, Fixer, and Reviewer. The Reproducer converts the bug ticket, logs, and specification into a failing test that proves the issue. The Investigator identifies the root cause and runs a pre-fix Blast Radius analysis. The Fixer patches the code until the tests pass. Finally, the Reviewer performs a post-fix Blast Radius check and prepares the pull-request description. The key feature is Blast Radius, our change-impact analyzer. A Python AST/tree-sitter scanner builds a call and dependency graph of the codebase. Bob then reasons over that graph to determine which dependencies represent meaningful risks. Rather than simply reporting that one function calls another, Bob can identify behavioral assumptions, such as a caller depending on an existing return type or behavior. Impacted components are classified as High, Medium, or Low risk with an explanation and information about test coverage. After the fix, the graph is generated again so the before-and-after results can highlight possible new regressions. For validation, Bug Squad uses an original sample application with deliberately seeded bugs and can also use closed issues and known fix commits from permissively licensed open-source repositories. These real fixes provide ground truth for comparing Blast Radius predictions with what maintainers actually changed. A lightweight dashboard is designed to show metrics such as time saved, tests added, and Blast Radius precision and recall. Bug Squad combines deterministic static analysis with Bob's contextual reasoning. Instead of using AI only to generate a patch, we use Bob throughout a structured engineering workflow: reproduce the problem, investigate its cause, implement the fix, test it, review it, and understand its wider impact.

VoiceWaiter - A Restaurant Menu You Can Talk To VoiceWaiter is an AI-powered voice-first restaurant ordering system that turns a traditional menu into a natural conversation. Instead of scrolling through menus or remembering exact dish names, customers can simply tell the AI what they are craving. For example: “I want something warm and filling under RM20.” The AI understands the customer’s intent, searches verified restaurant menu data, recommends dishes, handles dietary and allergy considerations and manages the cart through natural conversation. Customers can also browse the menu and order manually whenever they prefer. Key Features - 🎙️ Natural Voice Ordering — Order through everyday conversation. - 🔎 Smart Menu Discovery — Find dishes based on cravings, budget, dietary preferences and spice level. - 🛒 Voice Cart Management — Add, modify, remove and review items through voice or touch. - ⚠️ Allergy & Preference Handling — Clarify dietary requirements and communicate modifications. - 💰 Order Confirmation — Review items and total before placing the order. - 👨‍🍳 Live Staff Dashboard — Confirmed orders appear in a staff queue with real-time status updates. - 🍽️ What Should I Eat? — Get personalized recommendations when you don’t know what to order. How It Works 1. Scan → Customer scans the restaurant QR code. 2. Speak → Customer describes what they want naturally. 3. Discover → AI searches the verified menu and recommends suitable dishes. 4. Customize → Customer modifies their order through voice or touch. 5. Confirm → Cart and total are reviewed before submission. 6. Serve → The confirmed order appears in the staff dashboard. The AI searches verified restaurant menu data instead of inventing information. Menu items, prices, availability, allergens and order totals remain controlled by the restaurant’s data. Our vision: Turn the traditional static menu into an interactive culinary conversation, making ordering faster, more natural and more accessible.

💻 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

🌎 Online Hackathon | 💻 Month-long challenge | Dates: Sep 1–30, 2026 | 🏆 $10,000 Prize Pool ($5k cash + $5k in AAI credits) A month-long online challenge run by lablab.ai together with AssemblyAI, the Voice AI infrastructure company for builders. Every participant builds on AssemblyAI. You can join at any point — registration stays open for the whole build window, so you can start on day one or pick the project up halfway through and still submit. 🌍 Fully online hackathon. Join and build from anywhere in the world.