Hafiz Atif Ali@hmat
1
Events attended
1
Submissions made
Pakistan
5+ years of experience
About me
I am Hafiz Atif Ali, an AI engineer, full-stack developer, and founder of HMAT Technologies, where I build automation-driven products across healthcare, AI content systems, and enterprise software. My work focuses on solving real-world problems using practical, production-ready engineering. I specialize in Python, Node.js, Java, Next.js, Express, Prisma, and modern AI models, with strong skills in system design, LLM automation, computer vision, OCR, and deep learning. I enjoy building end-to-end intelligent pipelines—from idea to architecture to deployment. My key projects include: • Digital Health Record System: HIPAA-aligned, AI-powered platform that digitizes patient records using OCR and structured longitudinal data. • Brain Tumor MRI Analysis Pipeline: Research-grade system with EfficientNet, DenseNet, ResNet, ViT, and Grad-CAM/SHAP visualizations for IEEE research. • AI Animation Automation Engine: Converts user queries into full Python-generated animations with synced audio using LLM-powered prompt and code agents. • Solar Marketplace Platform: Modular Next.js + Node.js app with multi-role workflows, clean architecture, and production-level scalability. I love learning fundamentals deeply — including DSA, algorithms, OS, networking, and building data structures from scratch. My approach blends solid CS foundations with modern AI engineering. I aim to create accessible, intelligent automation tools that improve productivity, healthcare, and digital workflows. I'm passionate about turning ideas into real, working products and collaborating with teams that push what’s possible with AI.
.png&w=640&q=75)
🤓 Latest Submissions
QuantumDEX
QuantumDEX is an AI-powered decentralized exchange built natively on the Qubic Network to stop value leakage in DeFi trading. Today, users routinely overpay because swap routers check only a limited set of paths and ignore fragmented liquidity across chains, pools, and fee tiers. Our idea is to treat every trade as an optimization problem instead of a single DEX call. QuantumDEX runs machine-learning models on top of Qubic’s computational layer to simulate and evaluate thousands of potential routes in parallel before a transaction is sent on-chain. For each swap, we predict slippage, gas impact and MEV risk, then construct the route that maximizes the user’s expected output while honoring their constraints on speed and risk. Under the hood, QuantumDEX consumes on-chain and mempool data, updates models continuously, and exposes the intelligence through a simple, non-custodial interface and a developer-friendly routing API. Traders just connect a wallet and swap; integrators can plug our smart router into wallets, aggregators, or DeFi protocols. By combining scalable compute with adaptive ML, QuantumDEX aims to become the execution layer for Qubic-based DeFi, increasing capital efficiency for liquidity providers and giving end-users consistently better prices across all available pools.
7 Dec 2025
