Hamza Atiq@Hamza_Atiq

Lumi is a domain fine-tuned AI voice companion for dementia and Alzheimer's care. Built on AMD MI300X using QLoRA and GRPO reinforcement learning, Lumi handles confusion, repetition, and emotional fragility the way a trained caregiver would — not a generic chatbot. 55 million people live with dementia worldwide. Families cannot provide 24/7 care — and existing AI companions fail them. They reset every session, correct temporal confusion (which is clinically harmful), and leave patients vulnerable to scams costing $3 billion annually in elder fraud. Lumi is the first AI companion purpose-built for this population. Fine-tuned on AMD MI300X using QLoRA and GRPO reinforcement learning — the same technique behind DeepSeek-R1 — Lumi was trained on 8,540 dementia-specific samples processed through our EQ-Matrix framework, covering scenarios across severity levels, emotional states, and scam patterns. Persistent memory via ChromaDB injects prior session context into every new conversation — patients never repeat themselves. A structured output format fires the opening spoken line to TTS before the full response is generated, achieving time-to-first-audio under 1.5 seconds. A binary scam deflection classifier intercepts fraud attempts gently, without alarming the patient. On EQ-Bench 3, Lumi ranked 7th out of 46 models with a Rubric Score of 14.55 — confirming genuine emotional intelligence gains, not just surface fluency. The entire pipeline runs locally on AMD MI300X with ROCm and vLLM. No data leaves the device. No proprietary APIs. Fully private, fully open hardware.

"Agents shouldn't shop. They should settle." ASM is the first open protocol that gives AI agents structured data to evaluate, compare, and select AI services — then route a sub-cent USDC payment to the winner via Circle Gateway x402 on Arc testnet, in one HTTP call. THE PROBLEM When an autonomous agent needs to call an API — translate text, generate an image, transcribe audio — it picks between providers with zero structured data. Result: blind selection, 3–10× cost overrun, non-reproducible decisions. A data problem, not a model problem. THE SOLUTION (3 LAYERS) 1. DISCOVER — Registry of 70 manifests across 47 taxonomies (LLM, image, video, TTS, embed, GPU, DB). Each declares pricing, quality, SLA, and on-chain payment address. 2. EVALUATE — TOPSIS multi-criteria engine ranks candidates by cost × quality × speed × reliability. 72% top-1 taxonomy accuracy. 3. PAY — Each /api/score call resolves a winner and settles $0.005 USDC directly to that provider's Arc address via Circle Gateway. One endpoint, N recipients, sub-second finality. REQUIREMENTS — ALL MET ✅ Per-action ≤ $0.01: $0.005 avg ✅ 50+ on-chain tx: 50/50 settled, 0 failed, 15 unique recipients ✅ Margin: off-chain authorize + on-chain batch settle. Same 50 on L1 = $25–$250 gas. ~5,000× overhead eliminated. ✅ Stack: Arc Testnet (eip155:5042002) + USDC + Circle Gateway GOOGLE TRACK Gemini 2.5 Flash drives the routing loop via Function Calling. Agent receives a task + 30 taxonomies and emits a structured select_taxonomy_and_score call that triggers Circle x402 settlement. 80% accuracy. PARTNER STACK AI/ML API as Provider 0 (google/gemma-3-27b-it) for reranking and intent parsing; Gemini Native as Provider 1. OpenAI-compatible drop-in. LIVE Demo: asm-arc-circle-2026.vercel.app Code: github.com/calebguo007/asm-arc-circle-2026 On-chain: testnet.arcscan.app/address/0xF5d426D5cdfaeB18Ea2cDec2F7c2CB88eEe6b038 MIT.

Personalized TutorBot is an AI-powered multi-agent educational assistant designed to make AI and programming learning accessible, engaging, and personalized. Built using Blackbox-powered LLMs, FastAPI, and React, TutorBot intelligently guides learners through a customized journey via a chatbot interface. The system begins by understanding user goals and skill levels through natural conversation. Based on this, it dynamically generates personalized roadmaps, tailored tutorials, hands-on coding exercises, and intelligent code reviews—adapting in real time to learner progress. TutorBot includes key components like secure user authentication, an intuitive chatbot UI, an AI roadmap agent for goal-based planning, a tutorial agent delivering daily lessons, and a coding exercise agent that adjusts challenge levels. An integrated Replit-based IDE allows learners to code and test within the platform. The review agent offers instant, context-aware feedback on submitted code. All progress is tracked through a visual dashboard, helping users monitor their growth. With zero cost, zero setup time, and 100% personalized learning, TutorBot breaks financial and geographic barriers, scaling easily to serve thousands. Future enhancements include voice support for accessibility, gamified learning with XP and leaderboards, and an AI progress coach to offer proactive feedback and learning path adjustments—redefining how AI and coding education is delivered.

Auto-DevOps Agent is an AI-powered automation system designed to simplify and accelerate the setup of CI/CD pipelines for developers. Built for hackathons and real-world projects alike, it leverages a multi-agent architecture where each agent performs a specific DevOps task—detecting the tech stack, generating test/build/deploy steps, and combining them into a valid GitHub Actions YAML workflow. It also provides easy-to-understand explanations for each CI/CD step, enabling developers to learn while automating. The system supports various project types such as Python, Node.js, and Java, and is designed to eliminate the manual, error-prone work often required when configuring CI/CD tools. The agents communicate using structured prompts and are powered by LLMs via the Groq-hosted OpenAI SDK using the LLaMA 3.1-8B model. The full stack includes: Frontend: React.js (for user interaction and input handling) Backend: FastAPI (serving the AI agents and managing request logic) Agents: AI multi-agent system using prompt engineering and OpenAI SDK CI/CD Output: GitHub Actions YAML configuration Optional Extensions: Docker, GitHub Secrets, and platform deployment integrations The current version generates test, build, and deploy pipeline steps. Future enhancements will include direct deployment to platforms like Railway, Streamlit, Vercel, and Render, along with automated GitHub secret management and full repository initialization. This makes Auto-DevOps Agent an ideal solution for developers, students, and teams aiming to streamline DevOps without needing deep YAML or DevOps knowledge.

The AI Startup Builder & Pitch Generator is a multi-agent AI system that helps entrepreneurs validate, refine, and pitch their startup ideas. It automates market research, business model generation, and pitch deck creation, streamlining the startup journey with AI-driven insights. Core Features Idea Validation & Market Research Uses News API and Google Trends for real-time insights. Conducts competitor analysis and suggests improvements. Business Model Generation Develops revenue models and profitability estimates. Creates an AI-generated business plan. Pitch Deck Creation & Document Automation Generates investor-ready presentations with python-pptx. Exports structured reports using ReportLab. Provides elevator pitches and executive summaries. Technology Stack Streamlit – Interactive, multi-tab web application. News API & Google Trends (PyTrends) – Real-time market validation. LangChain & Community Modules – AI-powered text processing. OpenAI API – AI-driven content generation. CrewAI – Multi-agent collaboration. Impact & Target Audience Automates business planning, saving time and effort. Provides investor-ready, AI-backed insights. Ideal for entrepreneurs, first-time founders, and hackathon participants. Future Enhancements AI-driven legal and compliance assistance. AI-powered branding and marketing support. Investor matching for funding opportunities. AI-generated pitch videos for enhanced presentations. Investor Q&A coaching to prepare founders for discussions. This platform simplifies startup development by turning ideas into validated business opportunities, making AI entrepreneurship more accessible.

📅 May 11–19, 2026 • May 11 – 19: Online Build Phase. • May 18 (Hybrid Build Day – Online & Onsite) • May 19 (Demos & Awards – Online & Onsite) 🌍 Hybrid event. Join online from anywhere or build onsite at the venue. 🏆 Winners will be announced live on stage at the AI & Big Data Expo North America on May 19 📍 On-site Venue (May 18–19): San Jose McEnery Convention Center, CA, USA 🤝 Join solo or with a team. 🎟 FREE AI & Big Data Expo North America ticket included for participants that are able to join the on-site phase. 🏆Compete for a $10,000 prize pool 🧑‍💻 Apply now to design, prototype, and showcase enterprise-ready innovation.

Grab your ideas, spin up compute, and start building what’s next! Start with $100 in AMD Developer Cloud credits and begin building right away. Get access to hands-on tutorials, workshops, and curated learning resources to help you build faster and go further. To unlock credits and join the hackathon, sign up for the AMD AI Developer Program. 🤝 Go solo or team up — ship something real. 💰 $21,500+ total prize pool & 1 × AMD Radeon™ AI PRO R9700 GPU up for grabs. 📍 On-site Venue (May 9-10): MindsDB SF AI Collective 3154 17th St, San Francisco, California, USA 📍On-site participation is by invitation only. Travel and accommodation expenses will not be covered. 🧑‍💻 Developers, builders, and innovators - bring your boldest ideas and let’s turn them into real AI systems on AMD.

In this hybrid hackathon powered by Arc and Circle, developers will build the next generation of Agentic economic applications using Circle’s Nanopayments infrastructure with settlement on Arc in USDC. For the first time, high-frequency, usage-based transactions between users, APIs, and AI agents can operate without gas overhead eroding margin. Arc provides stablecoin-native settlement with predictable, dollar-denominated USDC fees and deterministic, sub-second finality — enabling real-time coordination and economically viable per-action pricing at scale. 📅 April 20 – 26, 2026 • Apr 20 – 25 (Online): Submit projects by Apr 25. • Apr 25 (On-site Build – SF, CA): Teams refine projects. • Apr 26 (On-site Demos & Awards – SF, CA): Live pitches & winners announced. Get guidance from Arc and Circle experts through on-site support. 📍 On-site Venue (April 25-26): MindsDB SF AI Collective 3154 17th St, San Francisco, California, USA 🤝 Go solo or team up, it's your flow. 🏆 Compete for a $15,000+ Prize Pool. 📍 Please note: If approved for the on-site experience, you’ll also have the opportunity to showcase your project live. Travel and accommodation will not be covered. 🧑🏻‍💻 Secure your spot now – sign up before the Kick-Off Stream!