AI Agents & Automation

Radium AI provides a comprehensive monitoring and management solution for Robotic Process Automation (RPA) bots, leveraging AI to enhance operational efficiency. The platform aggregates bot-run details from various RPA instances, including UiPath, Automation Anywhere, and Blue Prism, into a central repository. Its powerful ML engine classifies bot errors, identifies root causes, and recommends appropriate actions, significantly reducing incident resolution times. Radium AI also features an in-built action library and a robust workflow engine for defining custom automated actions against bot failures. It integrates with ITSM systems like ServiceNow to auto-generate tickets, ensuring 24x7 digital worker support and providing a single pane of glass for unparalleled bot observability.

text-to-lora offers a reference implementation of Text-to-LoRA (T2L), a system designed for instant transformer adaptation. This tool leverages hypernetworks to efficiently adapt Large Language Models (LLMs) for various benchmark tasks. A key feature is its ability to perform these adaptations using only textual descriptions of the desired tasks as input, simplifying the process of fine-tuning LLMs. The project provides detailed instructions for installation, running demos, generating LoRAs from the command line, and evaluating generated LoRAs. It also includes comprehensive guides for both SFT (Supervised Fine-Tuning) and Reconstruction training, making it a valuable resource for researchers and developers working with LLM adaptation.

Sentient.io offers a comprehensive AI & Data platform designed to empower enterprises with intelligent solutions. The platform provides ready-made AI models for easy adoption, allowing businesses to quickly integrate AI capabilities into their operations. Additionally, Sentient.io delivers turnkey AI solutions, catering to the specific needs of enterprises embarking on their digitalization journey. The platform emphasizes security and aims to simplify the process of leveraging artificial intelligence for various business applications, making advanced AI accessible for enhanced decision-making and operational efficiency.

SwitchOn Inc. offers AI-powered visual inspection solutions for manufacturers, primarily through its DeepInspect platform. This system automates quality inspection by using computer vision models trained on production data to identify defects such such as surface anomalies, assembly errors, and packaging inconsistencies in real time. DeepInspect integrates seamlessly with existing production lines, supporting various industrial-grade hardware and cameras, and can inspect up to 1000 parts per minute. It is applicable across diverse industries including automotive, pharma, electronics, and FMCG, helping to reduce product wastage, improve efficiency, and maintain high brand quality. The system continuously learns from new data, improving detection accuracy over time while providing real-time reporting and cloud analytics.

Deeplayer AI serves as a complete AI solutions partner, guiding businesses through the AI landscape from initial consultation to full deployment. Their services encompass strategic AI consulting, end-to-end implementation of AI solutions, and comprehensive training programs to upskill teams. A core offering is the creation of intelligent AI agents designed to automate tasks, enhance customer experience, and boost operational efficiency. Deeplayer AI also develops AI-powered platforms, such as Shootia.fr for custom photo generation and Ancien.ai for advanced AI photo restoration, showcasing their capability in building specialized AI tools. They emphasize a data-centric approach, ensuring AI agents connect seamlessly with existing data.

TextRecognitionDataGenerator is an open-source synthetic data generator designed to create text image samples for training Optical Character Recognition (OCR) software. It allows users to generate custom datasets with various parameters, including different fonts, backgrounds, and text modifications like skewing, blurring, and distortion. The tool supports multiple languages, including non-latin scripts like Chinese and Japanese, and can generate images with handwritten text (experimental). Users can run it via CLI or as a Python module, offering flexibility for integration into training pipelines. It also provides a Docker image for easier deployment, eliminating the need for local installations.

TheoremExplainAgent (TEA) is an open-source AI system designed to generate video-based multimodal explanations for Large Language Model (LLM) theorem understanding. It produces long-form Manim videos that visually explain mathematical theorems, demonstrating a deep understanding of the subject matter. This approach helps to uncover reasoning flaws that might be hidden in text-only explanations. The tool provides a comprehensive codebase for researchers, including generation and evaluation scripts. It supports various LLM models for video generation and offers features like Retrieval Augmented Generation (RAG) for enhanced context. TheoremExplainAgent is intended for academic research, particularly in the fields of AI, natural language processing, and educational technology, to advance the capabilities of LLMs in explaining complex mathematical concepts.

EasyFunctionCall is a SaaS service designed to streamline the integration of external APIs with various AI models, including ChatGPT, OpenAI, Claude, Gemini, and Llama. It achieves this by converting OpenAPI and Swagger specifications directly into AI model function call parameters. This process significantly reduces the complexity typically associated with API integration, making it easier for developers to leverage AI capabilities. Furthermore, the service optimizes token usage, which can lead to substantial cost savings for users. By providing a simplified method for handling API specifications, EasyFunctionCall enhances efficiency and accessibility for AI model development and deployment.

HaloMate AI is an all-in-one AI workspace designed for professionals to build and manage custom AI assistant teams. It unifies multi-model workflows, allowing users to switch between or compare models like GPT, Claude, DeepSeek, and more mid-chat. Users can create specialized 'Mates' with independent memories for specific domains, ensuring context isolation and persistent learning. The AutoPilot feature enables Mates to autonomously research across the web, news, and academic journals, providing visual insights and structured deliverables. HaloMate supports transforming ideas into various formats, including web applications, slides, and graphics, with instant preview and export options for seamless integration into professional workflows.

Silverstream AI offers an API and infrastructure specifically designed for building, scaling, and monitoring custom web browsing AI agents. The platform aims to simplify the complexities of developing reliable web agents, providing developers with the necessary tools through a handful of API endpoints. A key differentiator is its commitment to high accuracy, guaranteeing 95% (two sigma) reliability for web agents, with a goal to reach 99%. Silverstream AI emphasizes an incremental rollout approach, suggesting agents first operate within internal enterprise domains before expanding to user-facing applications. It treats web pages as a universal interface for agents and focuses on understanding and mimicking behaviors rather than just actions, enabling powerful agentic implementations.

MGM (Mini-Gemini) is an official repository for "Mini-Gemini: Mining the Potential of Multi-modality Vision Language Models." This open-source framework supports a series of dense and Mixture-of-Experts (MoE) Large Language Models (LLMs) ranging from 2B to 34B parameters. It is designed to facilitate image understanding, reasoning, and generation concurrently. Built upon the LLaVA framework, MGM also supports LLaMA3-based models. Key features include dual vision encoders for low and high-resolution visual embeddings, patch info mining for detailed region analysis, and an LLM for integrating text with images for both comprehension and generation. The repository provides models, data, and scripts for training and evaluation, making it a comprehensive resource for researchers and developers in multimodal AI.

Contenda, operating under FSH Technologies, is an AI-native software company dedicated to public service. They specialize in building government software solutions for municipalities, schools, and nonprofits. Their offerings range from food services and HR management to contract management and fundraising platforms. Contenda provides custom AI solutions, combining human expertise with advanced AI technology to simplify work and address specific business needs, including the integration of LLM agents and scaling AI operations.

Mirascope is an open-source LLM anti-framework designed to simplify interaction with various large language models (LLMs) through a unified interface. It empowers developers to integrate LLM capabilities into their applications using Python and TypeScript. Key features include the ability to call LLMs with simple decorators, retrieve structured output using Pydantic models, and build sophisticated AI agents equipped with tools. Mirascope supports advanced functionalities such as streaming, asynchronous operations, and multi-turn conversations, making it a versatile solution for developing complex AI-driven applications. The project is structured as a monorepo, providing clear separation for its Python and TypeScript implementations, as well as documentation and examples.

InfinityFlow is an AI-native database specifically designed for large language model (LLM) applications, offering incredibly fast hybrid search capabilities. It supports a wide range of search types including dense embedding, sparse embedding, tensor, and full-text search, alongside filtering and various rerankers like RRF, weighted sum, and ColBERT. The database boasts impressive performance, achieving 0.1 milliseconds query latency and up to 15K QPS on million-scale vector datasets. It supports rich data types such as strings, numerics, and vectors. InfinityFlow is built for ease-of-use with an intuitive Python API and a single-binary architecture, eliminating dependencies and simplifying deployment. It is available for Linux, Windows (via WSL/WSL2), and MacOS.

MNN is a highly efficient and lightweight deep learning framework developed by Alibaba, optimized for inference and training of deep learning models on devices. It delivers industry-leading performance and has been integrated into over 30 Alibaba apps, covering more than 70 usage scenarios. MNN-LLM, built on the MNN engine, enables local deployment of large language models on mobile phones, PCs, and IoT devices, supporting models like Qianwen, Baichuan, and LLAMA. MNN-Diffusion provides a runtime solution for deploying stable diffusion models locally. Key features include its lightweight design, versatility in supporting various model formats (Tensorflow, Caffe, ONNX, Torchscripts) and architectures, and high performance achieved through optimized assembly code and GPU acceleration.

n-skills provides a curated marketplace for AI agent plugins, emphasizing a 'write once, run everywhere' philosophy. It supports various AI coding agents including Claude Code, GitHub Copilot, Google Gemini, OpenAI Codex, Factory Droid, and Cursor, by utilizing universal formats like SKILL.md and AGENTS.md, and the openskills installer. Developers can install skills via agent-specific commands or the universal openskills CLI. The platform features categories like workflow, tools, development, productivity, automation, and data, and offers a process for submitting high-quality, value-add skills for inclusion. It also includes an auto-sync mechanism to keep external skills updated from their source repositories.

PnP.ai is an AI-as-a-Service platform designed to provide plug-and-play, industry-specific AI solutions for small and medium-sized enterprises (SMEs). The platform focuses on making AI accessible and easy to integrate, allowing businesses to leverage artificial intelligence without extensive technical expertise. It offers tailored AI solutions across various industries, aiming to streamline operations, enhance decision-making, and drive growth for its users. PnP.ai positions itself as a practical tool for SMEs looking to adopt AI technologies efficiently and effectively.

pyannote-whisper is an open-source tool designed for automatic speech recognition (ASR) and speaker diarization, leveraging the capabilities of Whisper for transcription and pyannote.audio for identifying and separating speakers. This tool allows users to process audio files to generate transcripts that include speaker labels and timestamps, making it ideal for analyzing multi-speaker conversations. It supports both command-line usage for quick processing and Python integration for more complex, programmatic workflows. The project provides clear examples for installation and usage, including how to integrate it into a Python script to diarize text and even generate meeting summaries using external LLMs like ChatGPT.

SiliconCedars is a Lebanese Offshore Development Center focused on building an ecosystem for the future through specialized software development and AI/ML services. The company excels in EDA development for Analog, Digital, and Mixed Signals Design, alongside offering comprehensive software services in Data Analytics, Big Data, and Artificial Intelligence/Machine Learning. Their vision is to create a world-class hub of ever-evolving, future-focused professionals. SiliconCedars is dedicated to identifying talent and transforming potential into opportunity, leveraging skilled and experienced engineering teams with perfected workflows and methodologies. They also highlight Neumann, a strong decision-making AI data platform that combines artificial intelligence, data analytics, business intelligence, and cold data analysis, offering solutions for collecting, mapping, and visualizing data.

SmartGPT is an experimental program designed to empower Large Language Models (LLMs), specifically GPT-3.5 and GPT-4, to tackle complex tasks without direct user intervention. It achieves this by intelligently breaking down large problems into smaller, manageable sub-problems and leveraging a robust plugin system to gather information from the internet and other external sources. The tool emphasizes modularity, allowing users to compose 'Autos' for various project requirements, and flexibility through a single, configurable `config.yml` file. While still in its early stages, SmartGPT aims for consistency in results through dynamic action execution and static tool-chaining, offering an innovative approach to autonomous AI task completion.

BeyondAI, operating as Beyond Limits, provides enterprise artificial intelligence solutions specifically for industrial environments. Leveraging advanced neuro-symbolic AI, the platform combines machine learning, generative AI, and rule-based reasoning to support complex operational decision-making. It focuses on high-stakes industries like energy, manufacturing, and infrastructure, where safety, uptime, and compliance are critical. BeyondAI offers solutions like Operations Advisor for AI-powered decision support, Beyond Search for secure enterprise knowledge intelligence, and AI in a Box for on-premise enterprise AI infrastructure, ensuring explainable and production-ready autonomy.

EASY2DIGITAL aims to help individuals and businesses master automation skills, allowing them to automate time-consuming but important tasks in life. The platform offers extensive guides and resources across various domains including AI & Automation Coding, Marketing Strategy, Digital & eCommerce, Finance & Investment, Web3 & Blockchain, and Javascript & React. It focuses on leveraging robots and AI to break free from the 24/7 grind, and uniquely emphasizes enhancing AI output with human warmth. Users can find tutorials on building AI agents, deploying web apps, creating keyword extractors, and understanding complex financial concepts, all designed to streamline workflows and boost productivity.

The C3.ai Digital Transformation Institute is a research consortium focused on advancing the application of artificial intelligence across various sectors. Established in March 2020, the institute brings together leading scientists to conduct research and train practitioners in the Science of Digital Transformation. This interdisciplinary field operates at the intersection of AI, machine learning, cloud computing, internet of things, big data analytics, organizational behavior, public policy, and ethics. The consortium includes C3 AI, Microsoft Corporation, and several prominent universities and national laboratories, jointly managed by the University of California, Berkeley, and the University of Illinois Urbana-Champaign. It supports research through various programs, including colloquia, symposia, and workshops, and publishes newsletters and articles on emerging science and research findings.

NLP-Projects is a comprehensive open-source repository dedicated to Natural Language Processing. It provides a wide array of concepts and practical scripts covering fundamental and advanced NLP topics. Users can explore implementations for word2vec, sentence2vec, machine reading comprehension, dialog systems, and text classification. The collection also delves into pretrained language models like XLNet, BERT, ELMo, and GPT, alongside sequence labeling, information retrieval, information extraction, knowledge graphs, text generation, and network embedding. It serves as a valuable resource for understanding and implementing various NLP techniques, with some sections offering Chinese notes for deeper insights.