Projects

### Overview: Quant-Finance-Resources is a comprehensive collection of courses, articles, and resources designed to help beginners and professionals in the field of quantitative finance. The repository is intended for STEM graduates, focusing on STEM skills and mathematics. The courses cover topics such as numerical linear algebra, probability, reinforcement learning, market microstructures, and advanced financial derivatives. ### Key Features: - **Numerical Linear Algebra for Coders** by fast.ai - **Introduction to Probability** by MIT OCW - **Topics in Mathematics with Applications in Finance** by MIT Mathematics - **Machine Learning** by Tom Mitchell - **Introduction to Reinforcement Learning** by David Silver - **Quantopian Lectures** for Python and Statistics - **Global Financial Crisis** by Timothy F. Geithner - **Introduction to Market Microstructures** by Paul Besson - **Python for Finance** by Yves Hilpisch - **Accelerated C++** by Andrew Koenig and Barbara E. Moo - **Effective Modern C++** by Scott Meyers - **Statistics** by David Freedman, Robert Pisani, Roger Purves - **How to Lie with Statistics** by Darrell Huff - **Thinking Strategically** by Avinash Dixit and Barry Nalebuff - **Analysis of Financial Time Series** by Ruey S. Tsay - **Deep Learning** by Ian Goodfellow et al. - **Reinforcement Learning An Introduction** by Sutton and Barto - **Options, Futures, and other Deriv

Overview: LibFewShot is a comprehensive library designed to simplify few-shot learning, making it accessible and easy to use for developers. Key features: - Non-episodic methods (Fine-tuning based) - Meta-learning based methods - Metric-learning based methods Tech stack: Python Why it matters: LibFewShot provides a unified framework for various few-shot learning methods, making it easier for developers to experiment with and integrate different techniques into their projects.

Overview: chromem-go is an embeddable vector database for Go with a Chroma-like interface and zero third-party dependencies. It is designed for simplicity and performance, enabling the integration of retrieval augmented generation (RAG) and similar embeddings-based features into Go apps without the need for a separate database. Key features: - Embeds documents into a vector database. - Supports multiple embedding providers and models. - Provides a zero-dependency interface for easy integration. - In-memory storage with optional persistence. Tech stack: Go Why it matters: chromem-go simplifies the process of integrating embeddings-based features into Go applications by providing a simple, embeddable database interface that doesn't require running a separate database. This makes it ideal for use cases such as retrieval augmented generation (RAG), text and code search, recommendation systems, classification, and clustering. Repository: philippgille/chromem-go Primary language: Go GitHub stars: 1069 GitHub description: Embeddable vector database for Go with Chroma-like interface and zero third-party dependencies. In-memory with optional persistence. GitHub badges: [Go Reference](https://pkg.go.dev/github.com/philippgille/chromem-go.svg), [Build status](https://github.com/philippgille/chromem-go/actions/workflows/go.yml/badge.svg), [Go Report Card](https://goreportcard.com/report/github.com/philippgille/chromem-go), [GitHub Releases](https://img.shields.io/github/release/philippgille/chromem

Overview: Dynamiq is an orchestration framework designed to streamline the development of AI-powered applications, specifically focusing on retrieval-augmented generation and large language model agents. Key features: - Dynamiq is an orchestration framework for agentic AI and LLM applications. - It specializes in orchestrating retrieval-augmented generation (RAG) and large language model (LLM) agents. - It provides a seamless integration of AI models and tools for developers. Tech stack: Python Why it matters: Dynamiq simplifies the development of AI applications by providing a unified platform for AI models and tools, making it easier for developers to create and deploy AI-powered applications.

Overview: MultiDiffusion is an official PyTorch implementation of the paper "MultiDiffusion: Fusing Diffusion Paths for Controlled Image Generation" presented at ICML 2023. It enables versatile and controllable image generation using a pre-trained text-to-image diffusion model without further training or fine-tuning. Key features: - Unified framework for versatile and controllable image generation. - Uses a pre-trained text-to-image diffusion model. - Supports localized and global edits, including texture changes and semi-transparent effects. - Available as a Jupyter Notebook repository. - Integrates with the diffusers library for seamless integration into existing projects. Tech stack: PyTorch. Why it matters: MultiDiffusion simplifies the process of generating high-quality, controllable images, making it accessible for various applications without the need for extensive re-training or fine-tuning.

Overview: Rogue is a comprehensive platform designed to evaluate AI agents and simulate adversarial attacks, making it essential for both security audits and penetration testing. Key features: - Automatic evaluation against business policies and expected behaviors - Red teaming to find security vulnerabilities - Multiple interfaces for client-server architecture - Support for various protocols including A2A, MCP, and Python Tech stack: Python Why it matters: Rogue provides a powerful tool for developers and security professionals to test and evaluate AI agents, ensuring they meet business requirements and security standards. Repository: rogue-security/rogue GitHub stars: 1071 GitHub description: AI Agent Evaluator & Red Team Platform README excerpt: # Rogue — AI Agent Evaluator & Red Team Platform  <div align="center"> <a href="https://trendshift.io/repositories/15191" target="_blank"><img src="https://trendshift.io/api/badge/repositories/15191" alt="qualifire-dev%2Frogue | Trendshift" style="width: 250px; height: 55px;" width="250" height="55"/></a>  <img src="./freddy-rogue.png" width="200"/> **Stress-test your AI agents before attackers do.** [Disc

Overview: `mcpc` is a universal MCP CLI client that supports a wide range of operations, including tools, resources, prompts, async tasks, skills, and more, making it the most compatible MCP client available. Key features: - Full MCP support - Persistent sessions - Progressive tool discovery - Code mode for JSON output - Secure OAuth 2.1 with DCR and CIMD - AI sandboxing for proxying MCP connections - Lightweight dependencies - Experimental x402 support for agentic payments Tech stack: TypeScript Why it matters: `mcpc` simplifies the use of MCP by mapping operations to intuitive shell commands, making it easier for AI agents to interact with MCP servers and achieve higher reliability and performance. Repository: apify/mcpc GitHub stars: 1067 GitHub description: The most compatible MCP CLI client. Supports: tools, resources, prompts, completions, async tasks, skills, stdio and streamable HTTP, persistent sessions, OAuth 2.1 with DCR, CIMD, client credentials, and ID-JAG, OS keychain, --json for code mode, grep for tool discovery, proxy for AI sandboxes, x402, and more. GitHub description excerpt: "Use it to manually inspect and debug MCP servers, to script repeatable MCP workflows in plain shell, or to give AI agents the full MCP protocol through a single `Bash()` tool call, so they can interact with any MCP server and its latest capabilities using the most universal programming interface there is:

### Overview: Lilac is a tool designed to enhance the quality of datasets used for training and fine-tuning Large Language Models (LLMs). It provides interactive exploration, curation, and quality control features to improve the efficiency and effectiveness of data preparation. ### Key Features: - **Interactive Exploration**: Utilizes LLM-powered search, filtering, and annotation capabilities. - **Data Curation**: Removes duplicates, PII, and obscure content to reduce dataset size and lower training costs. - **Collaborative Data Management**: Facilitates team collaboration on a single, centralized dataset. - **Time-Optimized Data Processing**: Supports cluster and title millions of documents with the power of LLMs. ### Tech Stack: Lilac is built on open-source technologies, including Python, Hugging Face Spaces, and various LLMs. ### Why It Matters: Lilac simplifies the process of preparing high-quality data for LLMs, making it more efficient and cost-effective. By leveraging LLMs for data exploration and curation, Lilac helps companies like Cohere and Databricks achieve better results in AI model training and fine-tuning.

### Overview: PixelLib is a versatile library designed for performing semantic and instance segmentation in images and videos. It supports PyTorch and Tensorflow, making it accessible for various operating systems. The library is built to enhance real-time computer vision applications, offering precise and efficient object segmentation. ### Key Features: - **Pytorch Version**: PixelLib Pytorch version uses PointRend for instance segmentation, providing accurate and real-time performance. It supports both Linux and Windows OS. - **Tensorflow Version**: PixelLib Tensorflow version is also available, offering alternative performance and accuracy. - **Semantic Segmentation**: PixelLib supports semantic segmentation, allowing you to identify and label different objects in an image. - **Instance Segmentation**: PixelLib supports instance segmentation, enabling you to segment individual objects within an image or video. - **Compatibility**: PixelLib is compatible with Python 3.7 and above, and supports PyTorch versions 1.6.0, 1.7.1, 1.8.0, and 1.9.0. ### Tech Stack: Python ### Why it Matters: PixelLib simplifies the process of performing object segmentation in images and videos, making it easier and more efficient for developers to integrate into their projects. Its advanced segmentation techniques and real-time performance make it a valuable tool for applications requiring accurate and timely visual analysis. ### Repository: ayoolaolafenwa/PixelLib ### GitHub Stars: 1068 ### GitHub Description: PixelLib is a library for performing segmentation of objects in images and videos. It

### Overview: Snow CLI is a powerful AI coding tool that allows developers to write code in the terminal, making it compatible with OpenAI, Gemini, and Claude. It runs on any terminal and is designed to be agentic, meaning it can seamlessly integrate with multiple AI models. ### Key features: - Snow CLI: The terminal-based interface for Snow CLI. - Supports multiple AI models: OpenAI, Gemini, and Claude. - Compatible with various AI models: Deepseek, OpenAI, Gemini, and Claude. - Agentic coding: Integrates with multiple AI agents for better collaboration and task delegation. ### Tech stack: TypeScript ### Why it matters: Snow CLI simplifies the process of AI-driven coding, making it accessible and efficient for developers. Its compatibility with multiple AI models and agentic nature make it a valuable tool for modern software development. ### Repository: MayDay-wpf/snow-cli ### GitHub stars: 1067 ### GitHub description: Agentic coding in your terminal, Simultaneously compatible with OpenAI, Gemini, and Claude. Running in the terminal, Snow CLI is designed to be agentic, allowing seamless integration with multiple AI models for better collaboration and task delegation. It supports Deepseek, OpenAI, Gemini, and Claude, making it a versatile tool for developers. The repository is maintained by MayDay-wpf, and it has been starred by over 1000 developers. ### Primary language: TypeScript ### GitHub description (English): Agentic coding in your terminal ### GitHub description (Chinese): 终端上的智能编程

### Overview: This repository is a collection of various ethical hacking tools and malware samples written in multiple programming languages, including Python, Ruby, Rust, C++, Go, and C. It serves as a resource for penetration testing and educational purposes. ### Key Features: - **Python Tools:** Scripts and libraries for various hacking and security tasks. - **Ruby Tools:** Scripts and frameworks for network scanning, vulnerability assessment, and more. - **Rust Tools:** Secure and efficient tools for low-level hacking and security analysis. - **C++ Tools:** Advanced tools for system exploitation, reverse engineering, and security research. - **Go Tools:** Comprehensive tools for network scanning, vulnerability assessment, and security analysis. - **C Tools:** Basic tools for system exploitation and security research. ### Tech Stack: - **Languages:** Python, Ruby, Rust, C++, Go, C. ### Why It Matters: This repository provides a comprehensive set of tools and samples for security researchers, ethical hackers, and educational purposes. It helps in understanding various hacking techniques and security vulnerabilities, contributing to the field of cybersecurity.

Overview: PyGame Learning Environment (PLE) is a reinforcement learning environment designed for quick experimentation in Python, inspired by the Arcade Learning Environment. Key features: - A comprehensive library of games, including Pong, Breakout, and more. - Simulates the Arcade Learning Environment interface for easy integration of existing agents. - Supports various game configurations and environments. - Provides a simple setup for beginners and advanced users alike. - Offers a headless usage option for testing without a graphical interface. Tech stack: Python, PyGame, Pillow, numpy, pygame. Why it matters: PLE simplifies the process of developing reinforcement learning models by abstracting away the complexities of game environments, allowing practitioners to focus on the design of their models and experiments.

Overview: Rust CV is a project aimed at implementing computer vision algorithms, abstractions, and systems in Rust. It contains pure-Rust dependencies that aim to encapsulate the capabilities of OpenCV, OpenMVG, and vSLAM frameworks. Key features: - Photogrammetry algorithms - Computer vision libraries - Computer vision utilities Tech stack: Pure-Rust dependencies Why it matters: Rust CV provides a cohesive set of APIs for computer vision tasks, focusing on algorithms and systems in Rust, while also maintaining compatibility with OpenCV, OpenMVG, and vSLAM frameworks. Repository: rust-cv/cv GitHub stars: 1068 GitHub description: Rust CV mono-repo. Contains pure-Rust dependencies which attempt to encapsulate the capability of OpenCV, OpenMVG, and vSLAM frameworks in a cohesive set of APIs. README excerpt: [...] Build instructions: Ensure Rust is installed. Install the necessary packages: Cmake, build-essential, freetype2, and libxkbcommon. Clone and build the repository using the following commands: ``` cd <directory to keep cloned repo> git clone https://github.com/rust-cv/cv.git cd cv cargo build ```

### Overview: StackQL is a powerful tool that enables humans and AI agents to query, provision, and operate resources across various cloud, SaaS, API, and Model Context Protocol (MCP) environments using a unified SQL-based framework. This allows for efficient and seamless management of complex systems. ### Key Features: - **Unified SQL-based Framework**: StackQL allows users to write SQL queries that can be executed against multiple types of cloud and SaaS resources. - **Resource Provisioning**: Automatically provision resources such as databases, APIs, and models without manual intervention. - **Model Context Protocol (MCP) Support**: Seamlessly integrate with Model Context Protocol for seamless integration with AI and machine learning tools. ### Tech Stack: Go ### Why It Matters: StackQL simplifies the management and querying of cloud, SaaS, API, and MCP resources, making it an essential tool for developers and data analysts who need to work with a variety of cloud environments. Its ability to automate provisioning and query across different types of resources makes it highly valuable for modern IT operations and data management.

Overview: OntoGPT is a Python package designed for extracting structured information from text using large language models (LLMs) and ontology-based grounding. It helps modern LLMs avoid hallucination and adhere to schema adherence without loading the entire schema. Key features: - Extracts structured information from text - Grounds names against ontologies - Supports multiple ontologies and models - Provides ready templates for various domains - Offers agent skills for AI agents to call OntoGPT Tech stack: Python 3.10 or greater, Jupyter Notebook Why it matters: OntoGPT simplifies the extraction process for LLMs, ensuring accurate and efficient information retrieval while avoiding common pitfalls like hallucination and schema adherence issues. Repository: monarch-initiative/ontogpt GitHub stars: 1067 GitHub description: LLM-based ontological extraction tools, including SPIRES

Overview: SearChat is an AI-powered conversational search engine built with TypeScript, supporting multiple search engines and deep research functionality. It integrates with OpenAI, Anthropic, Gemini, and SearXNG, offering a modern interface experience with responsive design, dark/light themes, and internationalization support. Key features: - Multi-model support for OpenAI, Anthropic, Gemini APIs - Multiple search engines integration - Conversational search with chat history - Deep research mode with structured reports - Support for SearXNG, Bing, Google, Tavily, Exa, Bocha, and ChatGLM Web Search Tech stack: TypeScript, Node.js + Koa backend, Vue 3 + TypeScript frontend Why it matters: SearChat is a versatile tool for anyone looking to enhance their search capabilities with AI-powered insights and a modern user interface.

### Overview: Power BI AI skills and Power BI agents for Claude Code and GitHub Copilot: a plugin marketplace for developers focused on Power BI and Microsoft Fabric. This marketplace offers a variety of skills, subagents, hooks, and plugins to enhance the performance of Power BI agents, including semantic models, reports, and workflows. ### Key Features: - **Power BI Skills**: Includes Power BI skills for data visualization, analysis, and reporting. - **Fabric Skills**: Features Fabric skills for developing and managing Power BI and Microsoft Fabric environments. - **Subagents**: Provides subagents for automating common tasks and workflows. - **Hooks**: Offers hooks for integrating with other tools and platforms. - **Reports**: Supports creating and managing Power BI reports. - **Semantic Models**: Enhances the semantic understanding of data within Power BI. - **DAX**: Provides DAX skills for advanced data analysis. - **TMDL**: Offers TMDL skills for creating and managing tabular models. - **Reports**: Supports creating and managing Power BI reports. - **Paginated Reports**: Enhances the pagination capabilities of reports. - **Custom Visuals**: Enables developers to create and use custom visuals in Power BI. - **Fabric CLI**: Provides command-line interface for managing Fabric environments. - **Fabric Admin**: Offers administrative functions for managing Fabric environments. - **Databricks CLI**: Enables integration with Databricks for data management. - **ETL**: Supports Extract, Transform, Load processes for data integration. - **Dashboards**: Facilitates the creation and management of Power BI dashboards

### Overview: This repository contains a curated list of academic papers focused on generative Information Extraction (IE) using Large Language Models (LLMs). The papers are organized into various tasks and techniques, providing a comprehensive overview of the field. ### Key features: - **Information Extraction Tasks:** - **Named Entity Recognition (NER):** - **Entity Typing:** Models that recognize and categorize specific entities. - **Entity Identification & Typing:** Models that identify and categorize entities. - **Information Extraction Techniques:** - **Supervised Fine-tuning:** Fine-tuning LLMs on labeled datasets. - **Few-shot:** Using a small number of labeled examples to train models. - **Zero-shot:** Models that can recognize entities without explicit training on the target domain. - **Data Augmentation:** Techniques to enhance the diversity of training data. - **Prompts Design:** Crafting specific prompts to guide LLMs towards the desired output. - **Constrained Decoding Generation:** Limiting the output to specific formats or structures. ### Tech stack: - Python, Jupyter Notebook, and various LLMs (e.g., GPT-3, Qwen, etc.). ### Why it matters: This repository serves as a valuable resource for researchers and practitioners interested in the latest advancements in generative Information Extraction using LLMs. It provides a structured way to explore and stay updated with the most relevant papers in the field. Repository: quqxui/Awesome-LLM4IE-Papers GitHub stars:

Overview: This guide aims to keep developers updated with the latest news, papers, discussions, and tutorials in the rapidly evolving fields of AI, Machine Learning, Deep Learning, and Computer Vision. Key features: - Stay connected to local communities through forums and meetups. - Access to essential resources and updates from various countries. Tech stack: None specified. Why it matters: This guide is essential for developers looking to stay current with the latest trends and advancements in AI, Machine Learning, Deep Learning, and Computer Vision.

Overview: Experts.js simplifies the creation and deployment of OpenAI's Assistants and Tools, making it easier for developers to create advanced Multi AI Agent Systems with expanded memory and attention. Key features: - Simplifies managing Run objects - Introduces Assistants as Tools - Supports instructions up to 256,000 characters - Integrates with 128 tools - Utilizes Vector Store API for efficient file search Tech stack: JavaScript Why it matters: Experts.js streamlines the use of OpenAI's Assistants API, making it more accessible and easier to integrate into existing projects, thus enabling developers to create more complex and advanced AI systems.