Coding & Development

Libretto is a powerful tool designed for software developers to monitor, test, and optimize LLM prompts. It moves beyond manual checks by offering comprehensive monitoring that automatically flags potential errors and identifies where LLMs are failing. The platform jumpstarts the evaluation process by generating test sets from production traffic and creating evals to judge LLM performance. Developers can instantly try new prompts, models, and strategies, getting actionable results in seconds. Libretto also features Drift Detection, which daily tests prompts to ensure models maintain consistent performance, preventing unexpected changes. It integrates seamlessly with existing workflows, providing real-time intelligence on LLM usage, costs, and quality, enabling continuous improvement of AI applications.

Unet-Segmentation-Pytorch-Nest-of-Unets is an open-source project offering a comprehensive collection of Unet model implementations for image segmentation tasks using PyTorch. This tool provides various architectures, including the original Unet, RCNN-Unet, Attention Unet, RCNN-Attention Unet, and Nested Unet (UNet++). It is designed for developers and researchers working on biomedical image segmentation or other image analysis problems. The repository includes code for data loading, model definitions, metrics, and visualization, making it a valuable resource for experimenting with and applying different Unet-based segmentation models. Users can easily clone the repository, install dependencies, and configure data paths to run the models.

Auctor is an AI-native system of action designed to streamline and enhance software implementations. It provides a unified platform for managing the entire implementation lifecycle, from the initial pre-sales phase through to go-live and ongoing support. By bringing together all aspects of a project, Auctor aims to deliver implementations with greater precision, control, and repeatable success. The platform helps teams manage complex projects, ensuring that every software implementation meets its full potential and delivers maximum value. It is trusted by top teams to improve the quality and efficiency of their implementation processes.

semisup-learn is a Python framework designed for semi-supervised learning, enabling the use of scikit-learn classifiers with datasets that are only partially labeled. It features implementations of Contrastive Pessimistic Likelihood Estimation (CPLE), a 'safe' framework applicable to classifiers that can yield prediction probabilities, ensuring model performance isn't worse than supervised-only training. The framework also includes Self Learning (self-training) and a wrapper for Semi-Supervised Support Vector Machine (S3VM) for comparison. CPLE is noted for its general applicability, low memory footprint, and reliance only on assumptions made by the chosen classifier, though it has high computational complexity. The project is an early-stage research endeavor.

RocketPages is a no-code website builder designed for small businesses, enabling users to create and publish professional websites quickly and easily. The platform offers a wide range of customizable templates tailored for various industries, ensuring a professional look without any coding knowledge. Key features include free hosting, 1GB of free storage on the starter plan, unlimited contributors, and robust SEO tools like auto-generated sitemaps and customizable meta tags. RocketPages also provides a user-friendly interface, responsive design, and a blogging platform, making it a comprehensive solution for establishing an online presence.

TERBINE is developing STRATA, a next-generation mobility infrastructure platform designed for consumer, commercial, and governmental applications. This platform leverages AI/ML, IoT, and cloud computing to provide real-time orchestration and synchronization for intelligent machines, including electric vehicles, drones, delivery robots, and bipedal robots. STRATA aims to improve safety, efficiency, and unlock new functionalities by enabling these machines to interact seamlessly with each other and their physical environments. A key use case is accelerating EV adoption by applying STRATA as a supervisory layer above charging network management systems, addressing issues like the high percentage of inoperable public chargers. For fleet operators, STRATA can provide real-time synchronization between vehicles and equipment in the field.

YouCompleteMe is a powerful, open-source code-completion engine specifically designed for the Vim text editor. It offers fast, as-you-type, fuzzy-search capabilities for code completion, comprehension, and refactoring. The tool integrates several completion engines, including a clangd-based engine for C-family languages, Jedi for Python, OmniSharp-Roslyn for C#, Gopls for Go, TSServer for JavaScript/TypeScript, rust-analyzer for Rust, and jdt.ls for Java. It also supports the Language Server Protocol for broader language compatibility and an identifier-based engine for all programming languages. Beyond basic completion, YouCompleteMe provides advanced IDE-like features such as signature help, finding declarations/definitions/usages, interactive symbol search, type information display, documentation in preview windows, code formatting, and semantic renaming across files. It also includes diagnostic display features, showing warnings and errors in real-time without needing to save the file.

hyperparameter-optimization is an open-source project providing implementations of Bayesian hyperparameter optimization for machine learning algorithms. This tool allows users to explore different approaches to hyperparameter tuning, specifically focusing on gradient boosting machines. It includes Jupyter Notebooks demonstrating the application of Bayesian optimization with libraries like Hyperopt, and provides examples for plotting search results. The project is designed to help data scientists and machine learning engineers enhance the performance of their models by systematically finding optimal hyperparameters, making the optimization process more efficient and effective.

HyperLandmark is a free and open-source tool designed for real-time face landmark detection, primarily targeting mobile applications. It utilizes deep learning to accurately identify 106 facial landmark points, offering a detailed facial contour description. The tool is noted for its high accuracy, even in challenging lighting conditions, and its efficient, small model size (around 2MB for the tracking model), making it highly suitable for mobile integration. It also supports multi-face tracking and boasts fast processing speeds, with the Android version achieving 7ms per single face on a Qualcomm 820. The project provides both Android and Windows implementations, with the Android version based on deep learning and the Windows version on traditional SDM algorithms.

NNoM is a high-level inference Neural Network library specifically designed for microcontrollers, aiming to provide a lightweight, user-friendly, and flexible interface for fast deployment on MCUs. It allows users to deploy Keras models to NNoM models with a single line of code and supports complex neural network structures such as Inception, ResNet, DenseNet, and Octave Convolution. The library features high-performance backend selections, including an optimized CMSIS-NN/DSP for ARM-Cortex-M, and onboard evaluation tools like runtime analysis and confusion matrix. NNoM manages structure and memory, simplifying embedded AI development. It is released under the Apache License 2.0.

Facetorch App is a Python library designed for comprehensive facial analysis, available as a Hugging Face Space. It allows users to upload photos or use a webcam to detect faces, generate 3D facial landmarks, and analyze various facial attributes. The app provides detailed reports on detected facial expressions, action units, and emotion scores. It also includes capabilities for extracting facial embeddings and performing face recognition. This tool is particularly useful for developers and researchers in computer vision who require advanced facial analysis functionalities for their projects.

Hyperbolic is an open-access AI cloud platform designed for developers, researchers, and enterprises. It provides affordable on-demand GPUs for training, scaling, and serving AI models, offering H100 or H200 instances in under a minute. The platform features serverless inference for running state-of-the-art AI models at a fraction of legacy cloud costs, with full OpenAI API compatibility. Users can also secure reserved clusters for guaranteed capacity and dedicated endpoints for high-throughput inference. Hyperbolic aims to be an end-to-end infrastructure solution, supporting various models like Llama, Qwen, DeepSeek, SDXL, and Flux, and offering AI consulting services for fast-scaling teams.

Geocalc MCP is an AI-powered geospatial tool developed during the Agents-MCP-Hackathon, designed to execute various geo-calculations independently, without relying on external third-party APIs. This application offers core functionalities such as converting addresses into precise geographical coordinates, calculating distances between points, and planning optimal routes. Users can also visualize these calculations and routes on maps, and identify nearby points of interest. It provides a self-contained solution for geospatial computations, making it suitable for projects requiring independent geo-processing capabilities.

VideoLLaMA3-Image is an AI tool designed for processing images and text inputs to produce detailed descriptive or analytical responses. This Hugging Face Space application leverages frontier foundation models for advanced video understanding, allowing users to explore and test AI models for video analysis. While the current live website indicates a runtime error, its intended functionality is to provide insights and answers based on visual and textual data, making it valuable for research and development in AI and video processing. The tool is developed by Xin Li and is available under an Apache 2.0 license.

mlhelper is an open-source JavaScript library designed for machine learning tasks, built upon Node.js. It offers a comprehensive suite of algorithms and utilities, including core functionalities like matrix and vector operations essential for numerical computations in ML. Beyond basic math, mlhelper supports practical aspects such as file parsing, enabling easy data ingestion, and feature engineering for preparing data for models. It also incorporates data visualization tools, particularly for graphs like Decision Trees and logistic regression, aiding in understanding model behavior. The project aims to foster a richer ecosystem for machine learning development within the JavaScript environment.

Sapper is an agentic AI platform designed for accounts payable automation, aiming to achieve over 70% straight-through processing. It significantly reduces labor dependency and modernizes finance operations without requiring ERP disruption. Unlike traditional AP automation that flags discrepancies for human intervention, Sapper autonomously resolves exceptions, executes policy-based approvals, and matches invoices across ERP, PO, and supplier history. The platform only escalates issues below defined confidence thresholds, leading to a higher touchless rate and reduced cost per invoice. Sapper ensures full auditability, embedded risk controls, and offers flexible deployment options including Multi-Tenant SaaS, Dedicated Tenant, and VPC/On-Premise, all while maintaining enterprise-grade security with ISO 27001 and SOC 2 Type II certifications.

Gemini Voyager is a browser extension designed to significantly enhance the user experience with Google Gemini. It provides an intuitive interface for navigating and organizing AI conversations, addressing common pain points like scrolling through long chat histories. Key features include a visual timeline for instant jumps to any point in a conversation, and a robust folder system for organizing chats, effectively acting as a file system for AI interactions. Users can also save and reuse their best prompts in a 'Vault', ensuring consistency and efficiency. The extension supports various export formats like JSON, Markdown, and PDF, giving users full data sovereignty. Additional functionalities range from context-aware quote replies and message timestamps to visual effects, lossless watermark removal, and advanced features like Mermaid diagram rendering and experimental conversation forks.

PandasAI is an AI dashboard solution designed to turn data into actionable insights rapidly. It serves as a comprehensive tool for business intelligence, offering robust capabilities for data visualization and automated reporting. The platform aims to simplify data analysis, allowing users to quickly understand complex datasets and generate reports without extensive manual effort. By leveraging artificial intelligence, PandasAI streamlines the process of extracting value from data, making it an efficient solution for businesses looking to enhance their decision-making processes through data-driven strategies.

Code Arena provides a platform for developers to interact with and evaluate leading AI coding models. Users can build web applications and websites in real-time, simultaneously assessing the accuracy and logical coherence of the AI's output. The platform features 'Battle Mode' for anonymous side-by-side model comparison, 'Side by Side Mode' for direct model selection, and 'Direct Mode' for focused interaction with specific models. It leverages a Bradley-Terry rating system, similar to Elo, to rank models based on community feedback, ensuring leaderboards reflect real-world performance. Code Arena also supports open research by sharing anonymized voting data and conversation logs to advance AI development.

Introduction to Machine Learning with Python is a comprehensive open-source repository designed to accompany the book of the same name by Andreas Mueller and Sarah Guido. It provides all the notebooks and code examples used in the book, making it an invaluable resource for students and practitioners looking to learn machine learning with Python. The repository includes helper functions from the `mglearn` library for creating figures and datasets, and all necessary datasets are included, with the exception of `aclImdb`. Users can set up their environment using `conda` or `pip` to install required packages like `numpy`, `scipy`, `scikit-learn`, `matplotlib`, `pandas`, `pillow`, and `graphviz`. It also supports `nltk` and `spacy` for text processing chapters.

SoTA-Point-Cloud is a GitHub repository offering an extensive survey of deep learning techniques applied to 3D point clouds. Published in IEEE TPAMI 2020, this resource covers major tasks such as 3D shape classification, 3D object detection, and 3D point cloud segmentation. It provides comparative results on numerous publicly available datasets, including ModelNet, KITTI, and Semantic3D. The repository also offers insightful observations and outlines future research directions, making it an invaluable resource for researchers and practitioners in the field of 3D computer vision. The maintainers regularly update the page with new results and suggestions.

Mintplex Labs Inc is an AI tool development company dedicated to creating innovative AI solutions for a wide range of business applications. Their core mission is to empower businesses by providing custom AI tools designed to enhance efficiency, streamline operations, and improve overall processes. While specific features are not detailed on their public-facing pages, their focus is on developing and deploying AI technologies that can be tailored to meet unique organizational needs. This suggests a strong emphasis on custom development and integration, rather than off-the-shelf products, making them a potential partner for businesses looking to leverage AI for competitive advantage.

Object Detection Safari is a free, web-based tool designed for exploring object detection through an interactive interface. Users can search for specific objects within images by providing text prompts, or upload their own queries to find relevant images and objects. The tool delivers labeled results, offering options to refine searches for more precise outcomes. It serves as an excellent resource for individuals interested in learning about object detection, providing a hands-on experience for educational and fun exploration. Developed by MyScale, it operates as a Hugging Face Space, making it accessible for anyone to experiment with AI-powered image analysis.

Score Jacobian Chaining is a technique designed for analyzing the sensitivity of machine learning models. This tool is invaluable for AI researchers and machine learning engineers seeking to understand the intricate relationship between model inputs and outputs. By providing insights into how changes in input data propagate through a model, it facilitates effective debugging and optimization. This understanding is crucial for improving model performance, ensuring robustness, and gaining deeper insights into model behavior. While the current live website indicates a runtime error, the underlying concept is highly relevant for academic research and practical application in machine learning development.