Flux helps users find products through natural conversation. Ask about laptops, headphones, or any product - the assistant searches real inventory and displays results alongside the chat. The interface splits between conversation and live product cards, creating an interactive shopping experience. ## Technical Implementation ### Core Stack - **LLM**: Groq API with `llama-3.3-70b-versatile` for tool calling - **Speech**: Groq Whisper (`whisper-large-v3-turbo`) for voice transcription - **Vision**: Groq Vision (`llama-3.2-11b-vision-preview`) for image analysis - **Video Avatar**: Tavus API for real-time AI video conversations - **Search**: Tavily API for product discovery - **Framework**: Next.js 15 with App Router - **Styling**: Tailwind CSS v4 ### Key Features **Multi-Agent System** Three specialized agents handle different domains: - Marketplace agent for product searches - Travel agent for flight and hotel queries - Food agent for restaurant recommendations **Real-Time Product Search** When users mention products, the system: 1. Extracts search intent using the 70B model's tool calling 2. Queries Tavily for real product data 3. Displays results in the interactive feed **Context Extraction** All conversations are parsed to extract: - Dates, locations, prices - Product names, flight numbers - User preferences and requirements This structured data appears in the activity feed, providing visual confirmation of what the AI understood.

ProcessLens is a comprehensive process mining and analysis platform built with IBM Granite 3.8B at its core. Our team developed a dual-layer analysis pipeline that processes operational data through both structured metric calculations and Granite's natural language capabilities. Technical Implementation: Frontend: Built with Next.js 15 and React 19, featuring real-time WebSocket updates and interactive visualizations using Recharts. The UI includes dark/light mode theming, responsive layouts, and enterprise-grade components from shadcn/ui. Backend: FastAPI implementation with a sophisticated analysis pipeline, MongoDB integration via motor, and GridFS for large file handling up to 100MB. Key Features: Data Processing & Analysis: Support for CSV and Excel with robust validation Parallel processing pipeline using IBM Granite for text analysis and structured metrics Real-time thought process visualization through WebSocket connections Comprehensive error handling with automatic retries Visualization & Metrics: Interactive performance dashboards showing cycle times and throughput Process pattern detection with confidence scoring Resource utilization analysis with heat maps Anomaly detection and bottleneck identification Enterprise Integration: WebSocket-based real-time updates MongoDB GridFS for scalable file storage Comprehensive API documentation PDF report generation with detailed insights Process Mining Capabilities: Automated pattern recognition in process flows Resource allocation optimization suggestions Performance bottleneck identification Cross-process correlation analysis Production Features: Role-based access control Dark/light mode support Responsive design for all devices Enterprise-grade error handling The system demonstrates advanced integration of IBM Granite through: Token-optimized prompting Structured output parsing Error recovery mechanisms Context-aware analysis

ConstructPlan uses AI to analyze project the shared data and provide actionable insights. By scheduling and resource management, it allows construction teams to focus on execution, minimizing delays and reducing costs. The platform’s intuitive interface ensures accessibility for all team members, regardless of technical expertise. Planning and Scheduling: Using data and AI, ConstructPlan generates optimized project timelines. The platform identifies potential bottlenecks in the project and offers recommendations to keep workflows on track. Automated scheduling adjusts task dependencies and manages milestones, providing real-time updates to avoid delays. Resource Management: ConstructPlan allows for precise allocation of personnel, equipment, and materials. With predictive analysis, it forecasts resource needs across the project lifecycle, reducing downtime and waste. The platform’s data-driven approach to resource management ensures optimal use of labor and materials. Risk Assessment and Mitigation: ConstructPlan provides AI-driven risk assessment by identifying and analyzing potential project risks. The platform continuously evaluates real-time data to highlight issues such as material shortages or labor shortages and suggests mitigation strategies before they affect the project’s timeline. AI-Driven Insights: By aggregating and analyzing project data, ConstructPlan enables managers to gain insights into progress and areas for improvement. These insights facilitate better decision-making and can prevent costly setbacks, ultimately leading to a smoother, more efficient project lifecycle. Document Generation and Reporting: With document generation features, ConstructPlan helps create comprehensive project reports and documentation. These AI-powered reports provide an accurate view of project status, compliance, and performance metrics, which can be shared with stakeholders, improving transparency.

RDJ (Reflect Discover Jarvis) is a web application that merges AI technology with psychological principles to create a unique platform for personal growth and self-discovery. The project utilizes advanced language models to generate personalized assessments, analyze user responses, and provide tailored insights. Users can explore their personality traits, emotional states, and potential areas for growth through interactive questionnaires and guided reflections. The application features a dynamic visual experience with AI-generated shader animations that respond to user input and mood, creating an immersive and engaging environment for self-exploration. AI chat support offers users guidance and clarification throughout their journey. RDJ aims to make professional-level psychological insights and personal development tools accessible to a wide audience, addressing the growing need for mental health and self-improvement resources in the digital age. The scalable architecture and continuous improvement cycle ensure that RDJ can evolve and adapt to user needs, potentially expanding into various sectors such as education, corporate wellness, and healthcare.

Granix is an innovative AI-powered project management platform that transforms how development teams work. By leveraging IBM's cutting-edge Granite AI models, Granix offers a suite of intelligent tools designed to optimize every aspect of the software development lifecycle. Key components of Granix include: Dashboard: The heart of Granix is a comprehensive dashboard that provides an at-a-glance view of project status, key metrics, and AI-generated insights. This centralized hub enables team members to quickly assess project health and make data-driven decisions. Code Structure Visualization: Granix features an interactive code structure tool that allows developers to easily navigate and understand complex project architectures. This visual representation of the codebase enhances collaboration and helps new team members quickly get up to speed. Task Management: Our intuitive task management system goes beyond simple to-do lists. Powered by AI, it helps prioritize tasks, estimates effort required, and suggests optimal task allocation based on team members' skills and workload. Project Overview: Granix provides a holistic view of multiple projects, allowing managers to efficiently oversee and coordinate various initiatives. The AI-driven insights help identify potential bottlenecks and suggest resource reallocation for optimal performance. AI Assistant: A chat-based AI helper offers instant, context-aware responses to project-related queries. This feature acts as a knowledgeable team member, providing quick answers and suggestions to keep the project moving forward. Code Assistant: Leveraging the power of IBM's Granite models, the Code Assistant provides AI-powered suggestions for coding tasks. It helps developers write better code faster, offering contextual recommendations and best practices.

Project Description: KodaWorld is an innovative AI-powered educational platform that transforms learning into an engaging adventure. The platform combines cutting-edge AI technology with interactive learning environments to create a personalized and enjoyable educational experience for students across various subjects. Key Features: AI-Powered Tutoring: Utilizes AI models (Falcon 180b, Falcon 11b, OpenAI API) to provide personalized assistance and explanations. Interactive Learning Environments: Includes a Music Visualizer, Particle Game, Shape Puzzle, Numbers Game, and Space Exploration tools to engage students in hands-on learning experiences. Adaptive Curriculum: Dynamically adjusts learning paths based on student performance, interests, and goals. Gamification: Implements achievements, challenges, and a points system to boost engagement and motivation. Character-Guided Learning: Features animal companions (Wake, Levo, Mina, Ella, and Koda) that guide students through different subject areas.

🔧 Spend 10 days launching your startup! 🚀 Improve your prototype previously developed at hackathons. 🤜 Collaborate with peers or work independently throughout the Hackathon. 🤖 Gain complimentary access to learning resources about Generative AI models, AI Agents, and beyond. 🚀 Explore opportunities with GAIA Startup Accelerator for Hackathon victors. 🌍 Connect and innovate alongside a worldwide AI community!

⌚ Dive into a 48-hour journey of creativity. 🚀 Compete to craft an app supercharged by Llama 3. 🎓 Expert mentors will guide you every step of the way. 🏆 Win an opportunity to participate in the Lablab NEXT acceleration program, exciting cash prizes, credits, and gear from LlamaIndex, Together AI & Zilliz 🧑🏻‍💻 Sign up before the Kick-Off Stream to secure your spot!

🕒 15 days to complete the challenge. It's an amazing opportunity to: 🌟 Explore new AI capabilities 🤝 Collaborate and build something extraordinary. 👥 Take part alone or form a team with other participants. 🎓 Receive guidance from expert mentors. 🏆 Compete for $32,500 in Prizes and a Full-Time Job Opportunity! 🎓 Exclusive job offers for top participants. 🧑🏻‍💻 Sign up before the Kick-Off Stream to secure your spot!