VerifyPay.AI is an AI-driven blockchain payment automation system that enables secure, transparent, and intelligent settlements between buyers and suppliers. Built on the Arc Blockchain using USDC and Circle Wallet integration, it employs AI/ML APIs to verify shipment or service completion in real time. Once validated, smart contracts automatically release payments; otherwise, funds remain locked to prevent fraud or delays. The system also supports cross-chain USDC transfers via CCTP and Bridge Kit SDK, ensuring global scalability. VerifyPay.AI delivers a tamper-proof, trustless, and efficient payment infrastructure powered by Cloudflare Workers AI and ElevenLabs Voice AI.

Chatterly is an AI-powered language learning platform designed to provide an interactive and immersive learning experience. Using GPT-5, the system generates tailored lessons with sentences in the target language, offering pronunciation guides and explanations to help learners understand meaning and context. The platform utilizes Deepgram's text-to-speech (TTS) to demonstrate accurate pronunciation, allowing users to hear and repeat sentences. During practice, Deepgram's speech-to-text (ASR) transcribes user input, enabling real-time feedback on pronunciation accuracy. With gamified features like progress tracking and achievement badges, Chatterly offers an engaging and personalized learning journey.

Our project is an agent-based Travel and Tourism platform designed to revolutionize the way users plan and manage their trips. At its core is an intelligent planner agent powered by Meta’s LLaMA 4 Scout (17B-16E-Instruct) model running on the Groq API. This enables real-time, context-aware decision-making to suggest destinations, accommodations, and travel activities tailored to user preferences like budget, location, and dates. To ensure up-to-date and relevant recommendations, the platform uses Google Custom Search API (CSE) and SerpAPI for real-time web data crawling. It gathers dynamic content about trending destinations, hotel prices, local events, travel advisories, and restaurants, ensuring users receive fresh and relevant suggestions. Once a trip is finalized, the planner agent automatically organizes the entire itinerary, including day-wise plans, hotel check-ins, flights, and activity bookings. Through seamless integration with n8n, it triggers webhooks that add the full itinerary to the user's Google Calendar, eliminating manual entry and improving organization. The system also handles booking confirmations, real-time alerts, and itinerary updates. A user-friendly frontend allows conversational interaction with the planner agent, enabling users to view, confirm, or adjust travel plans visually and easily. By combining AI-powered intelligence, real-time data crawling, and workflow automation, our platform delivers a smart, seamless, and personalized travel planning experience — handling everything from destination discovery to calendar management and booking confirmations in one place.

Our project is an AI-driven startup copilot designed to help early-stage founders go from idea to investor-ready with minimal friction. The user begins by entering their startup idea. The system performs automated market research using Retrieval-Augmented Generation (RAG) techniques that gather insights from sources like Google Search, Crunchbase, Reddit, and other public platforms. It analyzes the market landscape, identifies existing competitors, highlights potential gaps, and suggests unique value propositions. Once the idea is validated, the app uses large language models (LLMs) to generate a concise business plan, including a mini pitch deck covering problem, solution, target market, and monetization strategy. The outputs are downloadable as PDF or PPT formats. The MVP Builder Assistant then helps the founder shape the technical foundation of their product. It suggests the tech stack, auto-generates user stories, drafts database schemas, and outlines a feature roadmap to accelerate development. Finally, the Funding Advisor module recommends relevant VCs, accelerators, and angel investors using live data scraping and RAG techniques. It even drafts personalized outreach emails to help the founder secure funding more effectively. This all-in-one platform is ideal for solo founders, student entrepreneurs, and first-time startup builders who need structured guidance and speed without hiring a full team.

Our project, Telemedicine_BOT, is a web-based chatbot designed to help users quickly find nearby hospitals, check the availability of doctors and emergency beds, and get health-related guidance. It is especially useful in rural areas where people often don’t have quick access to hospitals or doctor contact information during emergencies. The chatbot is powered by a smart backend using FastAPI, LangChain, Chainlit, and vector search with FAISS, making it responsive and intelligent in understanding user needs. Users just need a phone or computer with internet access to use it. This solution is designed by a team of software engineers working together to build a meaningful healthtech product. It not only saves critical time during emergencies but also reduces costs by avoiding unnecessary travel and wait times. In the future, we plan to add features like live video calls with doctors, image-based disease identification, and offline SMS alerts for low-internet areas. The project has strong real-world potential, especially in countries like Pakistan, where many communities still lack fast access to medical care. Our goal is to build a tool that can help save lives, improve healthcare reach, and create a reliable digital bridge between patients and hospitals.

Contract Analysis Assistant is an AI-driven tool designed to streamline the process of reviewing legal contracts. Powered by IBM Granite, this assistant analyzes contract documents to identify potential risks, compliance issues, ambiguous terms, and key clauses that may impact legal and business decisions. Users can upload a PDF contract, and the AI will perform a detailed analysis, highlighting areas of concern such as legal pitfalls, operational risks, and clauses that may lead to disputes. The tool ensures faster contract reviews, reduces manual effort, and provides actionable insights to help legal teams and businesses make informed decisions. With an intuitive Streamlit-based interface, users can fine-tune AI settings like token limit, sampling parameters, and creativity levels to customize their analysis. Whether you are a lawyer, business owner, or compliance officer, the Contract Analysis Assistant simplifies complex contract analysis, making legal document review more efficient and accessible.

The AI-Powered Disaster Response System scans social media in real-time using DeepSeek AI/ML to detect disaster alerts & categorize theem into 3 categories based on severity level Smart Chatbot with DeepSeek RAG answers emergency queries using past disaster reports & safety guidelines which uses embeddings and vector database Voice Recognition enables non-English speakers to get real-time support. Emergency Alerts & Preparedness Guides ensure quick & informed action and provide links to youtube videos. It provides guidelines, psychological support and preaparedness checklist. We are working on integrating live location, IoT device integration and decentralized communication network

LaunchReady AI is a cutting-edge AI-driven space mission assistant that enhances mission planning, resource optimization, and communication for astronauts and space agencies. Powered by multi-agent AI models (LLaMA 3.3 Versatile), it streamlines fuel & power allocation, real-time data transmission, and supply chain logistics to prevent costly inefficiencies. With AI-powered automation, LaunchReady AI ensures faster mission readiness, reduces human error, and optimizes decision-making in deep-space exploration. It’s built using Streamlit, Groq API, and cloud-based infrastructure to provide real-time mission support. A must-have AI tool for NASA, SpaceX, and private space firms revolutionizing interplanetary travel! 🚀

Elevate the realm of machine learning with "WebML Assist." This innovative project integrates the power of WebGPU and the capabilities of the "BabyAGI" framework to offer a seamless, high-speed experience in machine learning tasks. "WebML Assist" empowers users to build, train, and deploy AI models effortlessly, leveraging the parallel processing of GPUs for accelerated training. The platform intuitively guides users through data preprocessing, model architecture selection, and hyperparameter tuning, all while harnessing the performance boost of WebGPU. Experience the future of efficient and rapid machine learning with "WebML Assist." Technologies Used: WebGPU OpenAI APIs (GPT-3.5, GPT-4) BabyAGI Pinecone API (for task management) FineTuner.ai (for no-code AI components) Python (for backend) Redis (for data caching) Qdrant (for efficient vector similarity search) Generative Agents (for simulating human behavior). AWS SageMaker (for developing machine learning models quickly and easily build, train, and deploy). Reinforcement learning (is an area of machine learning concerned with how intelligent agents). Categories: Machine Learning AI-Assisted Task Management Benefits: "WebML Assist" brings together the capabilities of WebGPU and AI frameworks like "BabyAGI" to provide an all-encompassing solution for ML enthusiasts. Users can seamlessly transition from data preprocessing to model deployment while harnessing the GPU's power for faster training. The incorporation of AI agents ensures intelligent suggestions and efficient task management. By integrating AI, GPU acceleration, and user-friendly interfaces, "WebML Assist" empowers both novice and experienced ML practitioners to unlock the true potential of their projects, transforming the way AI models are built, trained, and deployed.

🚀💻 Be the first to build an AI App on Google's models! Hackathon on July 7-10. 🔬🌐 Try new Vertex AI features from Google Cloud Platform. 🤝🌍 Learn from AI leaders and connect with like-minded people. 🛠️📱 Build apps with the world's best AI tools! 💡🌍 Solve real-world problems with Generative AI models

🗓️ 48-hour AI Challenge - Build your own AI agent or agent simulation over the weekend 🦜 Power up your coding with the LangChain framework! 🤝 Connect with potential co-founders and mentors at the event

🔴 An advanced-level hackathon ⏱ The one-day challenge 🧑‍💻 Build with the next-gen technology 🏁 Complete your app by the end!