Stars
The end of web parsing. The beginning of scalable pixel-native search. link: https://pixelrag.ai/
TensorRT YOLO live object detection pipeline for real-time video streams with FFmpeg, GPU acceleration, and annotated output.
An AI Agent that analyzes a csv file by answering questions, plotting graphs, and applying modifications. Can be used locally or using an MCP server.
Lightning-fast Rust workflow engine for building scalable agentic automation pipelines.
A minimal implementation of Manifold-Constrained Hyper-Connections (mHC) proposed by Deepseek in January 2026 for attention and MLP layers
General plug-and-play inference library for Recursive Language Models (RLMs), supporting various sandboxes.
A Lightweight Highly scalable and Available URL Shortener designed with FastAPI, Redis, and DynamoDB deployed using AWS ECS Fargate
A Proof-of-concept (PoC) that checks if a resume is a best fit to a Job description or not. A deep learning model was designed using dual-encoders trained on a generated synthetic dataset.
A production-ready multi-tenant RAG as a Service (RaaS) orchestrator
API, CLI, and Web App for analyzing and finding a person's profile in 1000 social media \ websites
An API that exposes an AI model that diagnosis and provides treatments for teeth radiograph images. Built using FastAPI and docker. Deployed using Hugging Face Spaces.
An Arabic News Credibility Checking platform that involved finetuning an LLM, building an API, and hosting it in the cloud using Azure