anaslimem/CortexaDB
It is a simple, fast, and hard-durable embedded database designed specifically for AI agent memory. It provides a single-file-like experience (no server required) but with native support for vectors, graphs, and temporal search.
Software Engineer
Repositories
anaslimem/career-ops
AI-powered job search system built on Claude Code. 14 skill modes, Go dashboard, PDF generation, batch processing.
anaslimem/llm-gateway-core
A production-grade LLM gateway that abstracts multiple model providers, implements intelligent routing, caching, retries, and observability to deliver reliable, cost-aware LLM access.
anaslimem/AI-Assistant-RAG
This project implements an AI-driven assistant using LLMs (Large Language Models) and a web search integration to answer user queries.
anaslimem/cargo-diagnose
It is a tool that checks the health of your Rust project's dependencies. It looks at your Cargo.toml file and checks your dependencies
anaslimem/termy
A Lightweight & GPU-rendered terminal emulator, written in Rust. Termy starts instantly and stays fast. Cross-platform 🦀
anaslimem/agent-client-protocol
A protocol for connecting any editor to any agent
anaslimem/rust-sdk
Rust SDK for ACP clients and agents.
anaslimem/symposium
AI the Rust Way
anaslimem/zed
Code at the speed of thought – Zed is a high-performance, multiplayer code editor from the creators of Atom and Tree-sitter.
anaslimem/chonkie
🦛 CHONK docs with Chonkie ✨ — The lightweight ingestion library for fast, efficient and robust RAG pipelines
anaslimem/anaslimem
Config files for my GitHub profile.
anaslimem/qdrant
Qdrant - High-performance, massive-scale Vector Database and Vector Search Engine for the next generation of AI. Also available in the cloud https://cloud.qdrant.io/
anaslimem/Rust
All Algorithms implemented in Rust
anaslimem/USearch
Fast Open-Source Search & Clustering engine × for Vectors & Arbitrary Objects × in C++, C, Python, JavaScript, Rust, Java, Objective-C, Swift, C#, GoLang, and Wolfram 🔍
anaslimem/openresponses
anaslimem/confidence-resolver
Flag Resolvers for Confidence
anaslimem/qdrant-client
Python client for Qdrant vector search engine
anaslimem/tantivy
Tantivy is a full-text search engine library inspired by Apache Lucene and written in Rust
anaslimem/Memoria
Memory-safe resource management system built in Rust. It demonstrates advanced concepts such as ownership, enums with associated data, safe error handling, and automated testing.
anaslimem/edgequake
High-performance GraphRAG inspired from LightRag written in Rust
anaslimem/llama_index
LlamaIndex is the leading framework for building LLM-powered agents over your data.
anaslimem/Autonomous_research_assistant
Agentic AI research assistant that autonomously processes domain-specific documents and generates structured insights through multi-agent collaboration.
anaslimem/vsadhkar
Taking Ajr to the moon 🚀
anaslimem/VectorMind-AI
AI-powered application that allows you to chat with your documents. Upload PDFs, text files, or paste raw text, and get intelligent, context-aware answers from a powerful language model.
anaslimem/Utopiahire
Utopiahire is an all-in-one platform that simplifies resume rewriting, mock interviews, job matching, and developer activity analysis. Combining a Python AI service, a NestJS backend, and a React frontend, it helps users enhance their career prospects and streamline the job search process.
anaslimem/Optimizers-from-scratch
Training a tiny MLP without using autograd, with manual forward/backward propagation, multiple optimizers, and learning rate schedules.
anaslimem/TweetFlowETL
An ETL (Extract, Transform, Load) pipeline project that extracts tweets from the Twitter API, transforms the data, and loads it into Azure Blob Storage.
anaslimem/Cybergraph
CyberGraph is a Neo4j-based knowledge graph that models relationships between IPs, malware, exploits, and target systems. It enables analysts to detect correlated threats, shared vulnerabilities, and attack propagation patterns using graph queries.
anaslimem/CNN-from-scratch
Convolutional Neural Network (CNN) trained on MNIST, built completely from scratch in NumPy (no PyTorch / TensorFlow).