Projects

Overview: Workflow SDK enables developers to build durable, reliable, and observable apps and AI Agents using TypeScript. Key features: - TypeScript support for building scalable and maintainable applications - Observability features for monitoring and debugging - Integration with Vercel for seamless deployment Tech stack: TypeScript Why it matters: Workflow SDK simplifies the development process by providing robust observability and reliability, making it ideal for building reliable and scalable applications and AI Agents.

Overview: 这个项目旨在展示如何使用 ChatGPT API 来完成各种 NLP 任务和非 NLP 任务，特别是通过 Prompts 和 Chains 的组合来提升生产力。 Key features: - 文本润色与改错 - 文本翻译 - NLU 语料扩充 - NLU 语料清洗 - NLU 推理 - 将 ChatGPT 作为 NLU 模型进行训练 - 将 ChatGPT 作为 NLU 模型进行推理 - ChatGPT 输出图像 - ASCII Art - SVG 图形 - graphviz 图形 - mermaid 图形 - ChatGPT 陪你游戏 - 井字棋 - 五子棋游戏 - 你画我猜 Tech stack: Shell Why it matters: 通过使用合适的 Prompts 和 Chains，可以将 ChatGPT 作为一个神奇的生产力工具，处理各种复杂的任务，从而提升工作效率和创造力。 Repository: howl-anderson/unlocking-the-power-of-llms Primary language: Shell GitHub stars: 2468 GitHub description: 使用 Prompts 和 Chains 让 ChatGPT 成为神奇的生产力工具！ README excerpt: # 使用 Prompts 和 Chains 让 ChatGPT 成为神奇的生产力工具！ ChatGPT 诞生后，因其非常强大的又难以置信的的能力，得到了非常广泛的关注。用户将 ChatGPT 视作一种

### Overview: This repository provides the official implementations of **3D Gaussian Ray Tracing (3DGRT)** and **3D Gaussian Unscented Transform (3DGUT)**. Unlike traditional methods that rely on splatting, 3DGRT performs ray tracing of volumetric Gaussian particles instead, enabling support for distorted cameras with complex, time-dependent effects such as rolling shutters. 3DGRT requires dedicated ray-tracing hardware and remains slower than 3DGS. To mitigate this limitation, 3DGUT enables support for distorted cameras with complex, time-dependent effects within a rasterization framework, maintaining the efficiency of rasterization methods. By aligning the rendering formulations of 3DGRT and 3DGUT, the hybrid approach called **3DGRUT** is introduced, allowing for rendering primary rays via rasterization and secondary rays via ray tracing, combining the strengths of both methods for improved performance and flexibility. ### Key Features: - **3D Gaussian Ray Tracing (3DGRT)**: Performs ray tracing of volumetric Gaussian particles, enabling support for distorted cameras with complex, time-dependent effects. - **3D Gaussian Unscented Transform (3DGUT)**: Enables support for distorted cameras with complex, time-dependent effects within a rasterization framework, maintaining the efficiency of rasterization methods. - **3DGRUT Hybrid Approach**: Combines the strengths of 3DGRT and 3DGUT to render primary rays via rasterization and secondary rays via ray tracing, providing improved performance and flexibility. ### Tech Stack: Python ### Why It Matters

### Overview: Zeroboot is a sub-millisecond VM sandboxing solution designed for AI agents. It leverages copy-on-write (CoW) forking to ensure isolation and minimal overhead. ### Key features: - **Sub-millisecond Performance**: Achieves execution times of less than 1 millisecond. - **Copy-on-Write Forking**: Ensures isolation between different sandboxed environments. - **Real KVM VMs**: Each sandbox runs as a separate KVM virtual machine with hardware-enforced memory isolation. ### Tech stack: Rust ### Why it matters: Zeroboot provides a lightweight, secure, and efficient way to sandbox AI agents, enabling rapid prototyping and testing without the overhead of traditional VMs. ### Repository: - **Repository**: zerobootdev/zeroboot - **GitHub Stars**: 2467 - **GitHub Description**: Sub-millisecond VM sandboxes for AI agents via copy-on-write forking - **Primary Language**: Rust ### Docs: - **API Reference**: docs/API.md - **Deployment Guide**: docs/DEPLOYMENT.md - **Architecture**: docs/ARCHITECTURE.md ### Author: - **Created**: March 2026 - **Maintained**: Adam Miribyan ([@adammiribyan](https://github.com/adammiribyan)) ### Cited by: - *DeltaBox: Scaling Stateful AI Agents with Millisecond-Level Sandbox Checkpoint/Rollback*, arXiv:2605.227

Overview: Sophus is a C++ implementation of Lie groups commonly used in computer vision and robotics, including SO(2), SO(3), SE(2), and SE(3). Key features: - Support for 2D and 3D rotations and isometries (rigid body transformations) - Geometric concepts such as unit vectors, splines, image classes, and camera models - Complete rewrite for sophus2, including additional geometric concepts - Rust version for sophus-rs with experimental non-linear least squares optimization library Tech stack: C++ Why it matters: Sophus is essential for implementing geometric transformations and computations in computer vision and robotics, providing a robust and efficient solution for tasks such as camera pose estimation, image alignment, and 3D reconstruction.

### Overview: 本仓库专注于“架构”而非“代码”的开源知识库，汇集了真实热门系统的架构模板，并配以成为更好架构师的教程。涵盖系统地思考方法、架构图绘制技巧、十大核心架构模式以及数据与状态管理等多方面内容。 ### Key Features: - **教程**：系统地教你“怎么像架构师一样思考”。 - **模板**：真实热门系统的架构地图，只讲架构不讲语法。 - **案例**：把具体项目从 0 推到上线，再推到真实压力下。 ### Tech Stack: Vue ### Why It Matters: 在未来，优秀的开发者首先是一个会做架构判断的人，其次才是会写代码的人。架构思维是这个时代的核心技能，它帮助你从需求出发，清晰地思考系统设计，而不是仅仅关注具体的实现细节。 --- Repository: study8677/awesome-architecture

Overview: The SVL Simulator is a ROS/ROS2 multi-robot simulator designed for autonomous vehicles, featuring advanced features and a comprehensive documentation. Key features: - Multi-robot simulation - Autonomous vehicle capabilities - Realistic vehicle models - Advanced simulation tools Tech stack: C# Why it matters: The SVL Simulator is essential for researchers and developers looking to create and test autonomous vehicle systems, offering a robust platform for simulation and validation.

Overview: RasaGPT is the first headless LLM chatbot platform built on top of Rasa and Langchain, offering a reference implementation for integrating LLMs with Rasa and Telegram. Key features: - 📚 Resources: [https://rasagpt.dev](https://rasagpt.dev) - 🧑‍💻 Github: [https://github.com/paulpierre/RasaGPT](https://github.com/paulpierre/RasaGPT) - 🧙 Author: [@paulpierre](https://twitter.com/paulpierre) Tech stack: Python Why it matters: RasaGPT simplifies the integration of LLMs with Rasa and Telegram, making it easier to create and maintain chatbots without the complexities of setting up a full Rasa project.

Overview: 这个项目是一个为开源大模型提供统一后端接口的项目，支持多种模型如LLaMA、LLaMA-2、BLOOM、Falcon、Baichuan、Qwen、Xverse、SqlCoder、CodeLLaMA、ChatGLM、ChatGLM2、ChatGLM3等。 Key features: - 支持多种开源大模型 - 实现流式响应，提供打印机效果 - 实现文本嵌入模型，支持文档知识问答 - 支持大规模语言模型开发工具 `langchain` 的各类功能 - 只需简单修改环境变量即可将开源模型作为 `ChatGPT` 的替代模型 - 支持加载经过自行训练过的 `lora` 模型 - 支持 `vLLM` 推理加速和处理并发请求 Tech stack: Python Why it matters: 该项目为开发者提供了一个统一的后端接口，简化了对多种开源大模型的调用，提高了开发效率。

### Overview: Rip'n Fast. Fewer Tokens. Better Code. **The ripgrep of AI context. A map before your agent reads the repo — and a check on what it writes.** Ranked, deterministic call graph: what to touch, what it breaks, which tests to run. On the edit: blast radius, tests that reach it, twelve quality kinds reporting only what got worse, forgotten co-changes, fields read and written, names that resolve more than one way. **Just want to use it?** Install it with the one line below, then start each coding session by telling your agent to use it, for example: *"Use ripwire on this repo."* That is all most people need: the install also teaches your agent when to reach for each command. **Want every detail?** [The reference guide](#reference-guide) near the bottom covers install, commands, output format, exit codes and limits. You do not need it to get started. ### Key features: - **Ranked Call Graph**: Identifies what to touch and what it breaks. - **Tests to Run**: Ensures you run the right tests. - **Blast Radius**: Shows what changes affect the most tests. - **Quality Delta**: Reports only what changed, revealing lost opportunities. - **Signatures**: Labels every guess and loss, ensuring clarity. - **Paddle Out with a Map**: Provides a quick overview of the project. ### Tech stack: C++23 CLI + MCP server ### Why it matters: - **Reduced

### Overview: Este repositório é dedicado à comunidade de cientistas de dados, oferecendo um conjunto de recursos e informações para ajudar a estender o conhecimento sobre análise de dados, empresas envolvidas nesse campo, e termos técnicos em português. O objetivo é reunir e organizar conhecimento em um local único, tornando-o mais fácil de acessar e compartilhar. ### Key Features: - **Datasets**: Muitos datasets disponíveis para uso em estudos e aprendizagem. - **Termos Técnicos**: Dicionário de termos técnicos em português. - **Empresas**: Lista de empresas brasileiras envolvidas em análise de dados. - **Grupos**: Tópicos de grupos de interesse em ciência de dados. - **Metodologias Ágeis**: Ferramentas e técnicas para desenvolvimento ágil. - **Dados em Imagens**: Tópicos relacionados a análise de imagens em dados. - **Metodologias Ágeis**: Ferramentas e técnicas para desenvolvimento ágil. - **Datasets**: Muitos datasets disponíveis para uso em estudos e aprendizagem. - **Termos Técnicos**: Dicionário de termos técnicos em português. - **Empresas**: Lista de empresas brasileiras envolvidas em análise de dados. - **Grupos**: Tópicos de grupos de interesse em ciência de dados. - **Met

### Overview: DeepBot is a system-level AI assistant designed for enterprise workflows, integrating seamlessly with existing systems to automate complex tasks and provide a real working partner. ### Key Features: - **Parallel Task Processing**: Run multiple tasks simultaneously without interference. - **20+ Built-in Tools**: File operations, command execution, browser control, image generation, image/video analysis, document analysis, AI chat, cross-session messaging, web fetching, Feishu/WeChat/WeCom messaging, Feishu document operations, and more. - **Memory System**: Long-term memory for user preferences and important context. - **Scheduled Tasks**: Automate recurring work with cron-based scheduling. - **Skill Extensions**: Compose tools into reusable Skills for complex workflows. - **Security Controls**: Strict path whitelist to protect system access. - **Multi-Model Support**: Qwen, OpenAI, Claude, and more. - **External Integrations**: Connect with Feishu and other platforms for cross-platform interaction. ### Tech Stack: TypeScript ### Why it Matters: DeepBot is a system-level AI assistant that integrates deeply with existing enterprise systems, enabling AI to participate in day-to-day operations across departments through multi-Agent collaboration. It supports parallel task execution, scheduled tasks, and skill extensions, all protected by strict security controls. The integration with Feishu and other platforms allows for cross-platform interaction, making it a real working partner for enterprise workflows.

### Overview: Model quantization is a crucial technique for optimizing neural network performance and reducing resource consumption. This collection includes papers, benchmarks, and tools for various quantization methods, such as post-training quantization, quantization-aware training, and low-bit models. ### Key features: - Papers, benchmarks, and surveys on model quantization. - Covers binary, ternary, vector, lattice, and low-bit networks. - Includes post-training quantization, quantization-aware training, and low-precision training. - Provides resources for vision, language, multimodal, generative, and state space models. ### Tech stack: Python, Markdown, Git ### Why it matters: Model quantization is essential for improving the efficiency and performance of neural networks, making them more suitable for resource-constrained environments.

Overview: lgo is an interactive Go programming environment that integrates seamlessly with Jupyter Notebooks, providing a powerful and intuitive way to write and execute Go code within a Jupyter notebook environment. Key features: - Write and execute Go code interactively like Python. - Jupyter Notebook integration. - Full Go language specification support. - Code completion and inspection in Jupyter Notebooks. - Display images, HTML, JavaScript, SVG, etc. - Currently only supported on Linux, but can be run on Mac and Windows with virtual machines or Docker. Tech stack: Go language, Jupyter Notebook, ZMQ, pkg-config. Why it matters: lgo offers a unique and powerful way to develop Go applications, especially for those who prefer the interactive and visual nature of Jupyter Notebooks. It simplifies the process of writing and testing Go code, making it an excellent choice for developers who frequently work with Go but need the benefits of a more modern development environment.

Overview: This repository is a curated list of community detection research papers and their implementations, focusing on Python. It includes papers and implementations related to various community detection techniques. Key features: - Papers and implementations of community detection methods - Python-based solutions for community detection - Comprehensive coverage of different community detection techniques Tech stack: Python Why it matters: This repository provides a valuable resource for researchers, developers, and practitioners interested in community detection algorithms and their implementations. It serves as a reference for understanding and applying various community detection techniques in Python.

Overview: Magic Copy is a Chrome extension that uses Meta's Segment Anything Model to extract a foreground object from an image and copy it to the clipboard. Key features: - Extracts foreground objects from images - Copies the extracted object to the clipboard - Supports installation via Chrome Web Store and manual installation Tech stack: TypeScript Why it matters: Magic Copy simplifies the process of extracting and copying foreground objects from images, making it easier for developers to integrate this functionality into their projects. Repository: kevmo314/magic-copy Primary language: TypeScript GitHub stars: 2454

### Overview: Metarank is a low-code Machine Learning personalized ranking service designed for articles, listings, search results, and recommendations. It boosts user engagement by integrating customer signals and using LLMs for real-time personalization. Metarank is accessible through a friendly Learn-to-Rank engine, making it easy to implement and scale. ### Key Features: - **Real-time Personalization**: Metarank adapts search results based on user actions and clicks. - **LLM-based Semantic Search**: Utilizes LLMs to understand the meaning of search queries. - **Stateless Cloud-Native Service**: Handles thousands of requests per second with minimal latency. - **Streamable Data Sources**: Integrates with various streaming processing systems for visitor signals. ### Tech Stack: Scala ### Why It Matters: Metarank simplifies the process of building smart search and recommendation systems, optimizing for high click-through rates and real-time personalization. With Metarank, developers can easily integrate customer signals and leverage LLMs for better query understanding, making their existing systems more effective and engaging.

Overview: Apify Agent Skills is a collection of web scraping and automation skills designed for AI coding agents. It provides a production-grade solution for scraping any site and building new Actors, making it easy to integrate into various AI platforms. Key features: - Scrape any site with built-in Actor selection across major platforms (Instagram, Facebook, TikTok, YouTube, X, LinkedIn, Google Maps, Reddit, Yelp, Airbnb, and more). - Build new Actors using the official SDK patterns in JavaScript, TypeScript, or Python. - Actorize existing code with proper input and output handling. - Integrate Apify into your product for workflow-automation apps, AI agent harness plugins, AI framework packages, or direct application clients via `apify-client` or the REST API. Tech stack: Python Why it matters: This skill set makes it easy to integrate AI agents into various applications, making them capable of scraping data and automating tasks across multiple platforms, enhancing their efficiency and effectiveness in data-driven workflows.

Overview: 这个项目是关于深度学习的实战教程，涵盖手写数字识别、Discuz验证码识别、垃圾分类和语义分割等多个应用场景。 Key features: - 手写数字识别 - Discuz验证码识别 - 垃圾分类 - 语义分割 Tech stack: Python Why it matters: 该项目提供了深度学习实战教程，涵盖了多个实际应用场景，帮助开发者理解和掌握深度学习技术。

Overview: TorchIO is a powerful library designed for medical imaging processing, leveraging deep learning techniques to enhance AI applications in the field of healthcare. Key features: - Advanced image processing tools - Comprehensive dataset management - Efficient model training and evaluation Tech stack: Python Why it matters: TorchIO serves as a crucial tool in the maturation of medical AI research, enabling researchers and developers to create more sophisticated and effective AI solutions for medical imaging. Repository: TorchIO-project/torchio Primary language: Python GitHub stars: 2451 GitHub description: Medical imaging processing for AI applications.