AI Agents & Automation

Deep-tempest extends the original gr-tempest project, also known as Van Eck Phreaking, by integrating deep learning techniques to significantly enhance the quality of spied images. This tool focuses on recovering visual information from unintended electromagnetic emanations, particularly those originating from HDMI cables. By applying advanced deep learning architectures like DRUNet, deep-tempest can reduce the Character Error Rate from 90% in the unmodified gr-tempest to less than 30%, making the recovered text much more legible. The project includes open-sourced code and a comprehensive dataset of synthetic and real captured images for research, training, and evaluation, supporting both Python 3.10 and 3.12 environments with Conda or Pyenv + venv setups.

Stable-Baselines3 (SB3) is a robust open-source library offering reliable implementations of reinforcement learning (RL) algorithms built on PyTorch. It serves as the next major version of Stable Baselines, aiming to facilitate the replication, refinement, and identification of new ideas within the RL community and industry. SB3 provides a common interface, supports custom environments and policies, and includes features like Tensorboard integration, custom callbacks, and high code coverage. While designed for ease of use, it assumes some prior knowledge of RL concepts. The library is actively maintained for bug fixes and documentation updates, with newer algorithms and faster variants developed in associated repositories like SB3 Contrib and SBX (SB3 + Jax).

Embedding Atlas is an open-source tool developed by Apple for interactive visualizations of large embeddings. It enables users to visualize, cross-filter, and search embeddings and associated metadata efficiently. Key features include automatic data clustering and labeling for interactive navigation of data structures, kernel density estimation and density contours to explore dense regions and outliers, and order-independent transparency for clear rendering of overlapping points. The tool also offers real-time search and nearest neighbors functionality to find similar data, and multi-coordinated views for metadata exploration. Built with WebGPU (with WebGL 2 fallback), it ensures fast performance for up to a few million points, making it suitable for data scientists and developers working with large datasets.

Daily Zaps is a leading newsletter platform dedicated to providing concise and timely updates on AI technology and news. Designed for busy professionals and curious minds, it distills complex AI developments into digestible summaries that can be read in approximately three minutes daily. The platform boasts a significant readership, including employees from prominent tech companies such as OpenAI, Apple, Hugging Face, and Adobe, underscoring its credibility and relevance in the AI landscape. Beyond its core newsletter, Daily Zaps also features an AI Tools Directory, helping users discover new AI applications, and an AI Jobs section, connecting talent with opportunities in the rapidly evolving AI industry. It serves as a valuable resource for staying informed without significant time investment.

tidybot2 is an open-source project providing a holonomic mobile manipulator designed for robot learning. It includes comprehensive hardware designs and software components for building and operating the robot. The platform supports various tasks, from phone teleoperation and data collection to policy training and inference. Its holonomic base allows for independent and simultaneous control of planar degrees of freedom, simplifying complex mobile manipulation tasks. The project offers a simulation environment for testing the codebase without physical hardware and detailed guides for assembly, usage, and software setup, making it accessible for researchers and developers in the field of robotics.

UAV_Obstacle_Avoiding_DRL is a comprehensive open-source project focused on developing deep reinforcement learning algorithms for autonomous obstacle avoidance in Unmanned Aerial Vehicles (UAVs). It addresses both static and dynamic environments, offering multiple approaches for each. For static environments, the project explores Multi-Agent Reinforcement Learning (MADDPG, DDPG, TD3) combined with artificial potential field algorithms. In dynamic settings, it utilizes disturbed flow field algorithms alongside single-agent reinforcement learning (PPO+GAE, TD3, DDPG, SAC). The project also includes implementations of traditional path planning methods like A* search, RRT, Ant Colony Algorithm, and D* algorithm for comparison, highlighting the superior performance of reinforcement learning approaches. It provides both MATLAB and Python implementations for various algorithms, making it a valuable resource for researchers and developers in UAV navigation.

legged_gym offers Isaac Gym environments specifically designed for training legged robots to walk on rough terrain. This open-source repository provides all necessary components for successful sim-to-real transfer, including an actuator network, friction and mass randomization, noisy observations, and random pushes during training. It supports the development and testing of robust robot control algorithms, with a focus on real-world applicability. The platform allows users to define and train new environments, customize robot assets, and fine-tune training parameters. While the project has migrated to Isaac Lab for future updates, this repository remains a valuable resource for those working with Isaac Gym Preview 3.

Awesome-Adaptation-of-Agentic-AI is a curated repository featuring a comprehensive list of academic papers focused on the adaptation strategies of agentic AI systems. This resource is designed for researchers and practitioners interested in the evolving field of agentic AI, offering insights into various adaptation methods. The repository categorizes papers based on agent adaptation (tool execution signaled, agent output signaled) and tool adaptation (agent-agnostic, agent-supervised), detailing development timelines, methods, venues, tasks, tools, agent backbones, and tuning techniques. It serves as a valuable reference for understanding the latest advancements and research trends in making AI agents more adaptive and intelligent.

MiniWorld is a minimalistic 3D interior environment simulator specifically designed for reinforcement learning and robotics research. It allows users to simulate environments featuring rooms, doors, hallways, and various objects, making it suitable for tasks like training AI agents in office, home, or maze-like settings. Written 100% in Python, MiniWorld is easily modifiable and extensible, offering features such as few dependencies, good performance, lightweight design, and support for domain randomization for sim-to-real transfer. It also provides fully observable top-down views, depth map production, and the ability to display alphanumeric strings on walls. This project has been deprecated as of August 11, 2025, and is no longer receiving updates or support.

gans-in-action is the official companion repository for the book 'GANs in Action: Deep Learning with Generative Adversarial Networks' by Jakub Langr and Vladimir Bok. This open-source resource allows users to reproduce, study, and extend every hands-on example from the book. It features Jupyter notebooks that walk through major variants in the GAN family, from vanilla GANs to CycleGANs, implemented using Keras/TensorFlow. The repository covers fundamental concepts of generative modeling, adversarial training, and best practices for stable GAN training. It includes implementations of architectures like DCGAN, Progressive GAN, Semi-Supervised GAN, and Conditional GAN, along with educational resources and canonical GAN papers.

everything-chatgpt is an open-source project hosted on GitHub that delves into the technical underpinnings of the ChatGPT web application. It offers a detailed exploration of various components, including backend API calls, data structures like session and user data, and model information. The project also covers beta features such as custom instructions, code interpreter, and plugins, along with how chat history and data export functionalities work. It's a valuable resource for developers, researchers, and anyone interested in understanding the technical architecture and operational mechanics of ChatGPT.

MyIP is a comprehensive, open-source IP Toolbox designed for detailed network analysis and diagnostics. It enables users to easily view their local and public IP addresses, perform IP geolocation lookups, and conduct essential network tests such as DNS leak detection and WebRTC connection examination. The tool also includes speed tests, ping tests, and MTR tests to assess network performance and connectivity. Additionally, MyIP offers website availability checks, WHOIS searches for domain and IP information, MAC lookups, and browser fingerprint analysis. It supports multiple languages, dark mode, a minimalist mobile-optimized mode, and PWA installation, making it a versatile solution for network professionals and users concerned with their online privacy and connectivity.

Dynamax is a Python package designed for probabilistic state space modeling, leveraging the JAX library for efficient computation. It offers robust capabilities for both inference (state estimation) and learning (parameter estimation) across a range of state space models. These include Hidden Markov Models (HMMs), Linear Gaussian State Space Models (Linear Dynamical Systems), Nonlinear Gaussian State Space Models, and Generalized Gaussian State Space Models with non-Gaussian emission models. The library provides both low-level, functionally pure inference algorithms and a user-friendly, object-oriented interface through model classes. It is compatible with other JAX ecosystem libraries like Optax for stochastic gradient descent and Blackjax for Hamiltonian Monte Carlo or sequential Monte Carlo.

InterpretML is an open-source Python package designed to bring clarity to machine learning models. It provides a unified framework for state-of-the-art interpretability techniques, enabling users to both train inherently interpretable 'glassbox' models like the Explainable Boosting Machine (EBM) and explain complex 'blackbox' systems using methods like SHAP and LIME. This tool is crucial for tasks such as model debugging, feature engineering, detecting fairness issues, and ensuring regulatory compliance in high-risk applications. It supports various data types natively and offers functionalities for global model understanding as well as explanations for individual predictions, making it a comprehensive solution for data scientists and machine learning engineers.

Introduction_to_Machine_Learning is a comprehensive GitHub repository offering educational materials for the Machine Learning course at Sharif University of Technology. This resource provides students and machine learning enthusiasts with access to detailed slides, interactive Jupyter notebooks for practical application, and various exercises to reinforce learning. Users can also find materials from previous semesters, ensuring a rich and evolving learning experience. Additional resources, including class videos, are available on the SharifML website (in Persian). The content is freely usable, with a request for proper citation of both the course and the GitHub repository under a Creative Commons BY license.

Gymnasium is an open-source Python library designed for developing and comparing reinforcement learning algorithms. It offers a standardized API for communication between learning algorithms and environments, alongside a comprehensive set of API-compliant environments. This library is a fork of OpenAI's Gym, maintained by the original team, ensuring continued development and support. It includes diverse environment families such as Classic Control, Box2D, Toy Text, MuJoCo, and Atari, catering to various complexity levels and problem types. Gymnasium also supports third-party environments and provides strict versioning for reproducibility. It is an essential tool for researchers and developers in the reinforcement learning field.

h1st offers power tools for AI engineers, championing a Human-First AI approach to address critical challenges in real-world data science. It helps overcome data scarcity in industrial AI by combining human knowledge with available data, enabling earlier time-to-market for intelligent systems. The platform fosters collaboration among data scientists by breaking down complex modeling problems into smaller, manageable parts, similar to established software engineering methodologies. Furthermore, h1st supports transparent and trustworthy AI by providing model description and explanation at multiple layers, which is crucial for deployment and regulatory compliance. It runs on Python 3.8 or above and can be easily installed via pip.

HybrIK is an open-source project offering a hybrid analytical-neural inverse kinematics (IK) solution for 3D human pose and shape estimation. It provides the official code for the research papers "HybrIK: A Hybrid Analytical-Neural Inverse Kinematics Solution for 3D Human Pose and Shape Estimation" (CVPR 2021) and "HybrIK-X: Hybrid Analytical-Neural Inverse Kinematics for Whole-body Mesh Recovery" (TPAMI 2025). The tool allows users to convert accurate 3D keypoints into parametric body meshes. Key features include demo code for visualizing HybrIK on videos and images, support for both SMPL and SMPL-X models, and a Blender add-on for importing results as FBX files. It also supports multi-person demos with pose-tracking and provides pretrained models with various backbones.

Inverse-Reinforcement-Learning is an open-source project providing implementations of various inverse reinforcement learning (IRL) algorithms. Developed as part of COMP3710, it was supervised by Dr Mayank Daswani and Dr Marcus Hutter. The project includes linear programming IRL (Ng & Russell, 2000), maximum entropy IRL (Ziebart et al., 2008), and deep maximum entropy IRL (Wulfmeier et al., 2015). Additionally, it features implementations of MDP domains like Gridworld (Sutton, 1998) and Objectworld (Levine et al., 2011). The repository also provides a final report detailing the implemented algorithms and offers module documentation for functions and classes.

machine-learning-and-simulation is a comprehensive GitHub repository offering handwritten notes and source code files that accompany YouTube videos on Machine Learning & Simulation. This resource caters to a broad audience, providing materials in both English and German. Key topics include foundational math for ML, essential probability functions, probabilistic machine learning (like VAEs and GANs), miscellaneous computer science topics, sparse matrices, continuum mechanics, automatic differentiation, Fenics tutorials, and various simulations implemented in Python or Julia. The repository also outlines future topics such as tensor calculus, ODEs, PDEs, and advanced machine learning techniques, making it a valuable learning hub for students and professionals alike.

Machine-Learning-for-Asset-Managers is an open-source GitHub repository offering practical implementations of code snippets and exercises from the book 'Machine Learning for Asset Managers' by Prof. Marcos López de Prado. This resource is designed for individuals looking to apply machine learning techniques to financial data, specifically within asset management. It covers topics such as denoising and detoning, distance metrics, optimal clustering, financial labeling methods like triple-barrier and trend-scanning, feature importance analysis, and portfolio construction techniques including Hierarchical Risk Parity (HRP) and Nested Clustered Optimization (NCO). The repository serves as a learning aid, allowing users to explore and replicate the book's concepts with real-world data.

Machine-Learning-From-Scratch is a GitHub repository by AssemblyAI-Community, offering implementations of various popular machine learning algorithms from scratch. This resource is designed to accompany AssemblyAI's Machine Learning from scratch course on YouTube, providing practical code examples for algorithms such as KNN, Linear Regression, Logistic Regression, Decision Trees, Random Forests, Naive Bayes, PCA, Perceptron, SVM, and KMeans. It's an excellent tool for students, developers, and data scientists looking to deepen their understanding of how these algorithms work at a fundamental level. The repository is based on a similar project by Python Engineer, ensuring a well-structured and educational approach to learning machine learning.

open-pi-zero is an open-source re-implementation of the pi0 vision-language-action (VLA) model from Physical Intelligence. This project aims to replicate the model's architecture, which adopts a Mixture-of-Experts (MoE) like design, where each expert has its own parameters and interacts through attention. The model integrates a pre-trained 3B PaliGemma VLM and a new set of action expert parameters (0.315B). It employs block-wise causal masking for efficient attention mechanisms and is trained using flow matching loss on the action chunk output. The repository provides installation instructions, details on testing with pre-trained weights, training specifics, and evaluation results, making it a valuable resource for researchers and developers in the field of VLA models.

SnakeFusion is an AI project that leverages genetic algorithms and neural networks to train virtual snakes within a game environment. The core concept involves training five individual snakes, which can then be fused together to create a single, more advanced 'ultimate snake'. This project serves as a practical demonstration of applying evolutionary algorithms and AI in game development. Built using Processing, it provides a hands-on approach to understanding how AI can learn and adapt. Users can interact with the system by adjusting mutation rates, saving trained snakes, and initiating the fusion process to observe the creation of a 'super snake'.