Coding & Development

Nidum.AI offers a decentralized and open-source AI ecosystem, empowering users to run AI models directly on their local machines. This platform eliminates the need for cloud infrastructure and intermediaries, making AI more accessible and transparent. Users can contribute their compute power to the network and, in return, earn rewards, fostering a collaborative and community-driven approach to AI development. Nidum.AI aims to democratize AI by providing a no-cloud, no-middlemen, no-limits environment for AI innovation and deployment. It's designed for those who want to leverage AI without relying on centralized services.

ValidMind is an agentic AI governance platform specifically designed for regulated financial services, including banking and insurance. It bridges the gap between rapid AI adoption and real-time regulatory compliance by offering a unified system for managing models, AI systems, and record types. Key capabilities include agentic AI governance, validation automation, and development automation, all designed to scale with agentic AI. The platform helps centralize oversight for responsible AI, unify AI and traditional model risk management, and automate documentation, testing, and workflows. ValidMind aims to accelerate AI adoption, reduce AI risk, and operationalize responsible AI, providing audit-ready evidence for various global regulatory standards.

Deepengine is a user-friendly platform designed to detect and manage vulnerabilities in both internal and external infrastructure. It helps companies find and resolve common vulnerabilities, misconfigurations, and outdated software to maintain a strong security posture. The platform offers consistent threat scanning, including attack surface management, infrastructure scanning for over 150,000 common vulnerabilities, and API & application scanning. Deepengine also provides internal vulnerability scanning through agents and simplifies compliance reporting with executive summaries. Based in Switzerland, it emphasizes data privacy and offers integrations with tools like Slack, Discord, and webhooks for enhanced threat visibility and response.

r4 Technologies delivers XEM (Cross Enterprise Management) software, a revolutionary AI solution designed to make organizations smarter, simpler, and more successful. XEM predicts demand to optimize supply across all business silos, leading to improved overall yield. This platform unifies disparate departments like marketing, sales, distribution, supply chain, operations, and people management, enabling them to work together more effectively. It is presented as a 'better way to AI' that doesn't require extensive infrastructure or data scientists, simplifying implementation. r4 Technologies applies this cross-enterprise AI across various industries, including commercial, public services, and federal applications, helping to uncover hidden revenue opportunities and improve decision-making.

pytorch-memonger is an open-source Python library designed for sublinear memory optimization in deep learning models built with PyTorch. It re-implements the technique described in "Training Deep Nets with Sublinear Memory Cost," significantly reducing the memory required for backward passes from O(N) to O(sqrt(N)). This optimization is particularly beneficial for training large sequential models or using larger batch sizes that would otherwise exceed available GPU memory. By replacing `nn.Sequential` with `memonger.SublinearSequential`, developers can easily integrate this memory-saving technique. The tool also addresses challenges with non-deterministic layers like BatchNorm and Dropout by re-scaling momentum and memorizing random number generator states, respectively, ensuring reliable performance.

safe-rlhf is an open-source framework designed for research into Constrained Value Alignment using Safe Reinforcement Learning from Human Feedback (RLHF). It offers a comprehensive and reproducible code pipeline, making it an invaluable resource for alignment research. The framework supports various training methods, including Supervised Fine-Tuning (SFT), standard Reinforcement Learning from Human Feedback (RLHF), and Safe RLHF. This allows researchers to explore different approaches to aligning AI models with human values while ensuring safety constraints are met. Its modular design facilitates experimentation and integration into existing research workflows, providing a robust platform for developing and evaluating safe AI systems.

shimmy is a Python-free Rust inference server designed for local AI model deployment, offering full OpenAI API compatibility. It supports GGUF and SafeTensors models, enabling users to point existing AI tools to shimmy for local, private, and free inference. Key features include hot model swapping, automatic model discovery from Hugging Face cache, Ollama, and local directories, and zero configuration. It also boasts advanced Mixture of Experts (MOE) support for running large models (70B+) on consumer hardware through intelligent CPU/GPU hybrid processing. shimmy is distributed as a single binary, including all GPU backends for various platforms, simplifying installation and ensuring automatic GPU detection.

uAvionix offers advanced avionics solutions for both crewed and uncrewed aircraft, with a strong emphasis on enhancing safety, communication, and navigation. Their comprehensive systems provide dual-frequency airspace visibility, integrating ADS-B traffic and non-cooperative crewed aviation traffic detection to support UAS operations. Key offerings include FlightLine for cooperative surveillance, CASIA G for detecting non-cooperative aircraft via ground-based computer vision, and Trakr for real-time UAS tracking. These technologies are designed to integrate seamlessly into modern aviation systems, empowering the future of flight with reliable and certified technology. The solutions are ideal for BVLOS operations, UAS dock operations, and public safety applications like Drone as a First Responder (DFR) and firefighting, ensuring aviation-grade visibility for all types of missions.

sunnypilot is an open-source driver assistance system, forked from comma.ai's openpilot, designed to provide a unique driving experience across more than 350 supported car makes and models. It achieves this by modifying behaviors of driving assist engagements, all while adhering as closely as possible to comma.ai's safety policies. The system logs various data, including road-facing camera, CAN, GPS, IMU, and thermal sensors, with optional opt-in for driver-facing camera and microphone. Users can join the community forum for support and installation instructions, and documentation is available for features and FAQs. sunnypilot is released under the MIT License, with portions derived from openpilot.

UniFace is a lightweight, production-ready, and open-source face analysis library built on ONNX Runtime, offering a comprehensive suite of features for processing facial data. It includes advanced capabilities such as RetinaFace, SCRFD, YOLOv5-Face, and YOLOv8-Face for face detection with 5-point landmarks, and various models like AdaFace, ArcFace, and MobileFace for face recognition embeddings. The library also provides multi-object face tracking with BYTETracker, 106-point facial landmark localization, and semantic face parsing with BiSeNet. Additional features include trimap-free portrait matting, real-time gaze and head pose estimation, and attribute analysis for age, gender, race, and emotion. For privacy, it offers anti-spoofing and face anonymization with multiple blur methods. UniFace supports hardware acceleration on ARM64 (Apple Silicon), CUDA (NVIDIA), and CPU, making it versatile for different deployment environments.

U-2-Net is an open-source project that provides an implementation of the U^2-Net architecture, specifically designed for salient object detection. This deep learning model employs a nested U-structure, which allows for more profound analysis of image features and significantly enhances its ability to accurately identify and segment salient objects. The architecture is known for achieving improved performance in various computer vision tasks related to object detection and segmentation. It is a valuable resource for researchers and engineers working in computer vision, offering a robust framework for developing and experimenting with advanced image processing techniques.

torch-points3d is a comprehensive PyTorch framework designed for deep learning on point clouds, offering a robust environment for researchers and developers. It facilitates the implementation and experimentation of deep learning models for various 3D data analysis tasks. The framework is built upon PyTorch Geometric and Facebook Hydra, ensuring reproducibility and ease of model construction. It supports a high-level API to democratize deep learning on point clouds and includes implementations of state-of-the-art models like PointNet, PointNet++, RandLA-Net, and KPConv. Users can leverage it for tasks such as classification, segmentation, object detection, panoptic segmentation, and registration, with support for datasets like ScanNet, S3DIS, and ModelNet.

Vehicle-detection is an open-source project focused on implementing vehicle detection using machine learning and computer vision. It leverages several key techniques, including Linear Support Vector Machines (SVM) for classification, Histogram of Oriented Gradients (HOG) for feature extraction, color space conversion for image processing, and a sliding window approach to scan images for vehicles. This project was originally developed as part of Udacity's Self-Driving Car Engineer Nanodegree, indicating its practical application in autonomous vehicle technology research and development. It serves as a valuable resource for students and developers interested in computer vision and self-driving car applications.

Windows-Machine-Learning is a comprehensive repository offering samples and tools for developing machine learning applications on the Windows platform. It provides a high-performance machine learning inference API, powered by ONNX Runtime and DirectML, suitable for low-latency applications like games and real-time systems. The repository includes various model samples for image classification and style transfer, as well as advanced scenario samples demonstrating custom tensorization, custom operators, and adapter selection. Developers can find tools for model conversion to ONNX format, optimization with WinML Dashboard and graph optimizations, and validation using WinMLRunner. It also features a Visual Studio extension (mlgen) for integrating WinML APIs into UWP apps, simplifying model loading and evaluation.

VLFeat is an open-source library designed for computer vision algorithms, specializing in image understanding and local feature extraction and matching. Written in C for efficiency and compatibility, it offers MATLAB interfaces for ease of use and detailed documentation. The library supports popular algorithms including Fisher Vector, VLAD, SIFT, MSER, k-means, hierarchical k-means, agglomerative information bottleneck, SLIC superpixels, quick shift superpixels, and large-scale SVM training. It is compatible with Windows, Mac OS X, and Linux, making it a versatile tool for developers and researchers in the computer vision domain.

ycmd is a powerful, open-source server designed to provide advanced code-completion and code-comprehension functionalities. Originally a core component of YouCompleteMe, it has been refactored into a standalone project, allowing its integration into various editors beyond Vim. The server offers APIs for semantic GoTo commands, diagnostic error and warning reporting, and intelligent code completion across multiple programming languages. It features several completion engines, including identifier-based, clangd-based for C-family languages, Jedi-based for Python, OmniSharp for C#, gopls for Go, TSServer for JavaScript/TypeScript, jdt.ls for Java, and RLS for Rust. ycmd intelligently merges results from different engines and supports user-level customization through a settings file and an extra configuration file for project-specific settings.

Keenly is a comprehensive identity theft and fraud protection service designed to safeguard your personal information online. It actively scans for your data across over 100 data brokers and automatically sends removal requests to delete your exposed information. The service includes continuous rescanning to ensure your data remains off these platforms, even if it reappears. Beyond data removal, Keenly offers a full privacy suite with dark web monitoring, neighborhood watch, and SSN monitoring to protect you and your family. It also provides $1 million in identity theft insurance to cover eligible losses and fees, alongside white-glove restoration services in case of identity compromise. Keenly aims to protect users from the growing risks of online crime and AI-facilitated identity theft.

Zinia is an AI platform designed to empower organizations in developing sophisticated AI solutions. The platform emphasizes a human-centered approach to AI, aiming to optimize business outcomes by making AI accessible and effective. It allows users to create AI models with minimal training and without requiring extensive technical expertise, catering to a broad range of industries and geographical locations. While the live website content is minimal, the stored description indicates its focus on simplifying AI development for diverse business needs.

BucléLabs is a pioneering blockchain development company dedicated to transforming digital landscapes through innovative solutions. They offer a wide array of services, including the development of blockchain frameworks like Parity Substrate, OP Stack, Arbitrum Orbit, Polygon zkEVM, and zkSync Hyperchains. Their expertise extends to Layer 1 & Layer 2 solutions for platforms such as BSC, Solana, Cardano, Ethereum, and Polkadot. BucléLabs also specializes in enterprise blockchain applications for supply chain, finance, healthcare, and real estate, alongside smart contract development and auditing. They provide tokenization services for assets and real estate, and white paper development, leveraging modern tech stacks like Laravel, Next.js, Python, and various cloud platforms.

Peak offers a cloud AI platform designed to help businesses build, run, and monetize AI applications. The platform provides agentic AI solutions that predict, decide, and act autonomously, specifically tailored for retail and manufacturing sectors. Key functionalities include commercial pricing optimization, inventory management across purchasing, production, and fulfillment, and intelligent merchandising automation. Peak aims to simplify AI adoption, offering pre-built products configured to specific business requirements for rapid time-to-value. The platform also supports connecting multiple products and use cases for integrated decision-making across an organization.

WiFi-3D-Fusion is an open-source research project designed to estimate 3D human pose by leveraging WiFi CSI (Channel State Information) signals and deep learning. It integrates wireless sensing with computer vision techniques to provide advanced spatial awareness. The system supports real-time motion detection and visualization, offering capabilities like multi-person 3D pose estimation, neural RF radiance fields, and 3D Wi-Fi scanning. It features robust CSI ingestion from local Wi-Fi sources, a real-time movement detector, and a 3D viewer. The project also includes a continuous learning system that automatically improves detection models and adapts to environmental characteristics, making it a powerful tool for researchers in wireless sensing and computer vision.

ASTER is an open-source attentional scene text recognizer designed to accurately recognize cropped text within natural images. It incorporates a flexible rectification mechanism to enhance recognition accuracy, particularly for challenging text orientations. The tool is implemented using TensorFlow r1.4 and reuses code from the TensorFlow Object Detection API, with a PyTorch port also available. ASTER provides scripts for data preparation, training, and on-the-fly evaluation, making it suitable for researchers and developers working on scene text recognition tasks. It includes a demo program with pretrained models for easy experimentation and offers state-of-the-art results in text recognition benchmarks.

Holocron Technologies specializes in developing AI-driven global domain awareness solutions for national security. The company leverages publicly available information to monitor and analyze advancements in science and technology worldwide. Their mission is to build sophisticated digital tools that ensure America's continued prevalence in the face of generational challenges. Holocron aims to provide critical insights and a "digital arsenal of democracy" by navigating unseen complexities and unraveling obscured information, ultimately building solutions for the defining challenges of our time.

ID Vision specializes in designing and integrating artificial vision systems with AI for demanding industrial production environments. The company provides turnkey solutions by combining its proprietary software, custom integration services, and expert support. Key offerings include VisionCloud, a cloud-based platform for training, continuous improvement, and remote management of AI vision models, and OneVision, a hybrid vision software that blends AI with classic vision algorithms for industrial quality control. ID Vision's solutions are designed to automate critical tasks such as intelligent quality control, real-time classification and sorting, and industrial process automation, including microscopic tasks and high-precision dimensional measurement. The company focuses on delivering reliable, scalable, and maintainable solutions to boost productivity and optimize operations through data-driven decision-making.