Qian Huang@Yatine

Our project is a multi-agent assistant for European public tenders that turns an RFP and a short company baseline into an actionable plan and a proposal draft. You provide an RFP link or file and complete a concise form with certifications, capabilities, financials, projects, and team skills. The system parses the RFP to identify deadlines, submission method and format, evaluation criteria and weights, eligibility requirements across legal, technical, and financial dimensions, mandatory attachments, and language constraints. It then compares those requirements to your baseline to produce a coverage matrix that marks items as covered, missing, or needs review, plus a focused to do list with evidence prompts. When coverage is sufficient, the writer agent generates Markdown sections such as Executive Summary, Technical Approach, Delivery and Timeline, and Quality and Risk, aligned to the scoring rubric and grounded in your capabilities. The MVP intentionally does not analyze existing proposal files; the goal is to minimize setup and guide teams to close gaps quickly. The architecture is modular and MCP compatible, so the same flow can run inside Coral or be exposed through simple APIs and a lightweight UI. Outputs can include anchors to source pages for easy verification and audit.

This project is an AI-powered personalized learning assistant built with Streamlit and deployed on Hugging Face Spaces. It helps users transform their learning goals into a structured study plan, using advanced large language models via the Aimlapi API. ✨ Key Features: Smart Questionnaire: Collects user background (e.g., beginner, student, IT professional), interests (Python, Java, C++…), and goals (job skills, academic research, hobby learning). Repository Recommendations: Based on the user profile, the system recommends high-quality GitHub repositories as learning resources. Learning Units: Each repository is split into study units derived from its README and documentation, making the content easier to digest. Interactive Q&A: On the study page, learners can ask questions in real time; the AI tutor answers patiently with explanations, examples, and practice problem suggestions (e.g., related LeetCode topics). Review Module: After completing a unit, the system generates structured summaries and saves them for later review, supporting spaced repetition. Seamless Deployment: Fully hosted on Hugging Face Spaces, requiring no installation—accessible from any device. This project is designed to be your patient and professional tutor, helping you not only learn from GitHub repositories but also apply the knowledge through guided practice and long-term retention.

Acumen AI addresses complex AI integration barriers preventing organizations from accessing advanced conversational AI due to technical complexity and expensive infrastructure. Our solution democratizes AI through an intelligent platform combining Groq's high-performance inference, Llama-3.3-70b-versatile model, MCP agents and Streamlit interface deployed on Vultr’s cloud server. The 4-layer architecture features asynchronous processing, persistent memory, and advanced error handling, enabling natural language conversations with 10-step reasoning capabilities. Users interact through a responsive web interface, initializing AI agents, engaging in real-time conversations, and managing memory without technical expertise. The Total Addressable Market (TAM) for conversational AI reaches $32.6 billion globally, with our Serviceable Addressable Market (SAM) targeting $8.4 billion in enterprise AI assistants. Revenue streams include SaaS subscriptions, API usage fees, premium features, white-label licensing, and consulting services. Unlike competitors such as ChatGPT Enterprise or Microsoft Copilot, Acumen AI offers open-source flexibility, superior MCP agent reasoning, cost-effective Groq infrastructure, and specialized conversational AI focus rather than general-purpose tools. Future prospects include horizontal scaling, knowledge base integration, multi-modal capabilities, and vertical-specific solutions, positioning Acumen AI as a foundational platform for next-generation accessible conversational systems.

We’ve built a thoughtful digital companion designed to help people escape the endless scroll of short videos not by blocking apps or setting strict limits, but by gently understanding and responding to your habits. Rather than just tracking how much time you spend on your phone, it pays attention to how you’re using it how quickly you scroll, how often you switch videos, the mood of the content you’re watching, and more. When it notices you might be stuck in a loop, it steps in with care. This companion doesn’t lecture or interrupt it starts a friendly, meaningful conversation. It might gently encourage you to take a short break, focus on a small goal, or just breathe and reflect for a moment. It’s not about control. It’s about support helping you feel more in charge of your time and attention, while still enjoying the things you love.

It’s a great opportunity to: 🚀 Leverage Google’s advanced Gemma 2 models to create cutting-edge solutions. 💰 Access resources to support your project development. 🌟 Receive guidance from expert mentors throughout the hackathon. 🤝 Work solo or in teams to bring your ideas to life. 🏆 Compete for a chance to choose from unique Google swag and enjoy an opportunity to meet the Gemma team for the best solutions! 🧑🏻‍💻 Sign up before the Kick-Off Stream to secure your spot! 🗓️ On-site Workshops: Join us in-person at Google Cloud Space in London on 28th November from 5:00 PM to 9:00 PM UK time (GMT) for hands-on workshops

🕒 Complete the challenge within 48 hours and unlock incredible prizes! 📅 June 13 - June 15 – Participate online, collaborating with innovators and builders to explore a new way of software development. 🌟 Expert mentors will guide you every step of the way. 🤝 Work alone or form a team to build something extraordinary. 💰 $80,000 in credits and prizes! ⏳ Sign up before the Kick-Off Stream to secure your spot!

This is your chance to push boundaries, solve real-world challenges, and create impact at the official hackathon of the RAISE Summit 2025 at one of Paris's most iconic venues: Le Carrousel du Louvre. 📅 July 4–8, 2025 — Collaborate online with a global community of innovators. 🏛️ July 8–9, 2025 — Limited on-site access to the RAISE Summit in Paris (300 participants), exclusively for selected teams. 🏆 July 9, 2025 — Winners will be announced live on stage during the RAISE Summit. 🌟 Expert mentors will guide you throughout your hackathon journey. 🤝 Work solo or build your dream team to create something extraordinary. 💰 Compete for a share of up to $150,000 in prizes. 🔒 Sign up before the Kick-Off Stream to secure your spot!