Projects

### Overview: PyRIT is an open-source framework designed to help security professionals and engineers identify risks in generative AI systems. ### Key Features: - **Risk Identification**: PyRIT helps in identifying potential risks in generative AI systems. - **Security Professionals**: Empowers security professionals to proactively identify risks. - **Engineers**: Supports engineers in understanding and mitigating risks in AI systems. ### Tech Stack: Python ### Why it Matters: PyRIT is crucial for security professionals and engineers working with generative AI systems, providing a tool to proactively identify risks and enhance the security posture of these systems.

### Overview: PilotDeck is a task-oriented AI Agent productivity platform designed to streamline and optimize workspaces for multi-project scenarios. It aims to redefine operational boundaries and enhance memory efficiency through a unified approach to productivity. ### Key Features: - **Visual Configuration**: Easily configure and visualize tasks with a simple, intuitive interface. - **Long-Running Tasks**: Handle and track long-running tasks across multiple interaction messages. - **Office-Ready Results**: Deliver high-quality Word, Excel, and PowerPoint skills with a single click. ### Tech Stack: TypeScript ### Why It Matters: PilotDeck simplifies complex AI workflows, making it a valuable tool for developers and teams looking to enhance their productivity and efficiency in the Agent era. Its ability to manage and optimize multiple projects in a unified environment makes it a compelling choice for modern workspaces. Repository: OpenBMB/PilotDeck Primary Language: TypeScript GitHub Stars: 4589 GitHub Description: Task-oriented AI Agent productivity platform GitHub Description: Refined by Tsinghua University, ModelBest, OpenBMB, and AI9Stars, PilotDeck is designed to be a true productivity tool for the Agent era. It addresses the challenges of managing multiple projects and tasks, offering a seamless and efficient workflow. The platform is built with TypeScript, making it a robust and reliable choice for developers and teams looking to streamline their workspaces.

Overview: Promptify is a task-based NLP engine built with Pydantic, designed for structured outputs, evaluation, and a unified LLM backend. It supports models like GPT-4o-mini and is compatible with Pydantic schemas for various NLP tasks. Key features: - Task-based NLP engine - Pydantic structured outputs - Built-in evaluation metrics - Supports Pydantic schemas for custom tasks - Compatible with various LLM models Tech stack: Python Why it matters: Promptify simplifies the process of applying LLMs to NLP tasks by providing a structured output format and built-in evaluation metrics, making it an ideal choice for researchers and developers looking to leverage LLMs for NLP projects. Repository: promptslab/Promptify GitHub stars: 4643 GitHub description: Prompt Engineering | Prompt Versioning | Use GPT or other prompt based models to get structured output. Join our discord for Prompt-Engineering, LLMs and other latest research.

### Overview: FastVideo is a unified inference and post-training framework designed for accelerated video generation, enabling real-time and post-processing tasks on various hardware platforms. ### Key Features: - **Unified Inference and Post-Training**: FastVideo simplifies the process of generating videos by integrating both inference and post-processing steps. - **Real-Time Generation**: Supports real-time video generation tasks, making it suitable for applications requiring immediate output. - **Post-Processing Enhancements**: Enhances the quality of generated videos through advanced post-processing techniques. - **Hardware Compatibility**: Works seamlessly with both CPU and GPU, including Apple Silicon and NVIDIA DGX Spark. ### Tech Stack: Python ### Why it Matters: FastVideo offers a streamlined solution for video generation, making it accessible and efficient across different computing environments, from desktops to cloud-based setups. Its ability to handle real-time and post-processing tasks efficiently makes it a valuable tool for developers and content creators alike.

### Overview: 《PyTorch实用教程》（第二版）是一本全面覆盖PyTorch基础知识、产业应用和推理部署的教程书籍。本书不仅适合零基础入门者，也适用于已经在深度学习领域有所建树的开发者。通过丰富的案例和详细的代码示例，读者可以轻松掌握PyTorch的使用，并将其应用于各种实际项目中。 ### Key features: - **基础模块**：从PyTorch的安装和环境搭建开始，逐步讲解模型、数据、优化和可视化等核心模块。 - **产业应用**：涵盖计算机视觉、自然语言处理和大语言模型等领域的实战案例，帮助读者将理论知识应用于实际项目中。 - **推理部署**：介绍推理部署框架ONNX和TensorRT，以及量化技术PTQ和QAT，帮助读者将模型从训练阶段转移到实际应用中。 ### Tech stack: - **Python**：使用Jupyter Notebook进行代码编写和展示。 - **PyTorch**：核心编程框架，用于构建和训练深度学习模型。 - **ONNX**：用于模型转换和部署，支持多种模型格式。 - **TensorRT**：高性能推理加速工具，适用于模型部署。 ### Why it matters: 本书不仅提供基础的PyTorch知识，还通过丰富的案例和实战项目，帮助读者掌握深度学习工程师所需的各种技能。无论是零基础入门，还是进阶应用，本书都能提供全面的指导和帮助，助力读者快速成长为优秀的深度学习工程师。 ### Repository: - **GitHub**: T

### Overview: Koog is a Kotlin-based framework designed to build and run AI agents across JVM, JS, WasmJS, Android, and iOS targets. It leverages Kotlin Multiplatform for multiplatform development and integrates with popular JVM frameworks like Spring Boot and Ktor. ### Key Features: - **Multiplatform Development**: Deploy agents across JVM, JS, WasmJS, Android, and iOS targets using Kotlin Multiplatform. - **Reliability and Fault-Tolerance**: Handle failures with built-in retries and restore the agent state at specific points during execution with the agent persistence feature. - **Intelligent History Compression**: Optimize token usage while maintaining context in long-running conversations using advanced built-in history compression techniques. - **Enterprise-Ready Integrations**: Utilize integration with popular JVM frameworks such as Spring Boot and Ktor to embed Koog into your applications. - **Observability with OpenTelemetry Exporters**: Monitor and debug applications with built-in support for popular observability providers (W&B, Weave, Langfuse). - **LLM Switching and Seamless History Adaptation**: Switch to a different LLM at any point without losing the existing conversation history, or reroute between multiple LLM providers. - **Integration with JVM and Kotlin Applications**: Build AI agents with an idiomatic, type-safe Kotlin DSL designed specifically for JVM and Kotlin developers. - **Model Context Protocol Integration**: Use Model Context Protocol (MCP) tools in AI agents. - **Agent Client Protocol Integration**: Build ACP-compliant agents that can communicate with standardized client applications using the Agent Client Protocol (ACP