AI Agents & Automation

Deeligence is an AI-powered platform designed to significantly accelerate due diligence and contract review processes, aiming to reduce human error and meet ambitious deadlines. It centralizes all due diligence projects and processes, providing a clear overview of progress. Key features include a Change Tracker for managing uploads and revisions, an AI Contract Screener that extracts over 100 contract fields with local law summaries, and an Early Warning System that uses agentic AI to identify and notify teams of red flags on day one. The tool also offers end-to-end solutions, data room agnosticism, team visibility, instant data import, one-touch reporting, and Q&A management, all while ensuring security and privacy with GDPR compliance and SOC-2/ISO 27001 in progress.

BudgetML is an open-source library designed for practitioners who need to quickly deploy machine learning models to an endpoint without significant time, money, or effort. It addresses the challenges of cloud functions' limitations and Kubernetes' overkill for single models by offering a simple, developer-friendly solution. BudgetML deploys models on Google Cloud Platform preemptible instances, which are approximately 80% cheaper than regular instances, while ensuring high uptime through automatic autostart. It provides features like automatic FastAPI server endpoint generation, interactive Swagger docs, built-in SSL certificate generation, and OAuth2 secured endpoints. While not intended for full-fledged production, it offers a cost-effective and fast way to get ML models into production.

NVIDIA Deep learning Dataset Synthesizer (NDDS) is a powerful UE4 plugin designed for computer vision researchers. It facilitates the export of high-quality synthetic images along with comprehensive metadata, including segmentation, depth, object pose, bounding boxes, keypoints, and custom stencils. Beyond simple export, NDDS incorporates various components for generating highly randomized images, encompassing lighting, objects, camera positions, poses, textures, and distractors, as well as camera path following. These capabilities collectively enable researchers to effortlessly create diverse and randomized scenes, which are crucial for effectively training deep neural networks and overcoming the limitations of hand-labeled data.

Daydreams is an AI agent framework designed for building agents, with a core focus on Agentic Commerce. It features a composable context architecture, allowing developers to create complex agents from modular components that remember, learn, and scale. The framework provides universal access to various AI models (OpenAI, Anthropic, Google, Groq, xAI) through its Dreams Router, supporting pay-per-use micropayments. Daydreams also offers native Model Context Protocol (MCP) integration, enabling agents to connect to external tools like file systems, databases, and web services. It is TypeScript-first, ensuring type safety and excellent developer experience, and supports persistent memory across sessions.

Dayflow is an open-source, local-first application for Mac that functions as an automatic work journal. It privately monitors your screen activity to create a chronological timeline of your day, helping you reconstruct what you actually did without manual notes or timers. The tool offers features like daily standup summaries, weekly reviews of focus patterns and app usage, and the ability to chat with your work journal to ask questions about your activities. Dayflow prioritizes privacy, storing all recordings and data locally on your Mac. Users can choose their preferred AI provider for analysis, including local models via Ollama or LM Studio, or cloud-based options like Gemini, ChatGPT, or Claude, balancing privacy, cost, and quality.

ElatoAI offers a comprehensive solution for integrating realtime voice AI into Arduino ESP32 devices, supporting over 100 voice AI models. It's designed for creating AI toys, companions, and various smart devices, facilitating uninterrupted conversations for more than 20 minutes globally. The platform leverages secure WebSockets and Deno Edge Functions for low-latency performance and global accessibility. Key features include real-time speech-to-speech conversion using APIs like OpenAI, Gemini, and Eleven Labs, custom AI agent creation with customizable voices, and robust hardware integration with the ESP32 Arduino Framework. It also provides device management, user authentication, conversation history, and OTA updates, making it a versatile tool for developers building interactive voice AI applications.

The fastapi-langgraph-agent-production-ready-template is a comprehensive solution for AI engineers looking to build robust AI agent backends using FastAPI and LangGraph. This template addresses critical aspects of AI agent development, including stateful conversations, long-term memory management, and tool calling. It integrates essential features like Langfuse tracing for observability, Prometheus metrics with Grafana dashboards for monitoring, and JWT authentication with session management for security. Additionally, it includes rate limiting via slowapi, Alembic migrations for database management, and an optional Valkey/Redis cache layer. The template is designed to handle the complex infrastructure, allowing developers to focus on agent logic and accelerate their application development.

faster-rnnlm is an open-source toolkit designed for efficient recurrent neural network language modeling. It aims to train on massive datasets (billions of words) and very large vocabularies (hundreds of thousands) for real-world Automatic Speech Recognition (ASR) and Machine Translation (MT) problems. The toolkit incorporates advanced setups like ReLU+DiagonalInitialization, GRU, Noise Contrastive Estimation (NCE), and RMSProp to achieve better results and faster training. It boasts impressive speed, processing over 250k words per second on a 3.3GHz CPU with standard parameters, making an epoch take less than an hour. The toolkit supports various hidden layer types and offers both Hierarchical Softmax and NCE for output layers, with NCE being particularly effective for large vocabularies as its speed is independent of vocabulary size.

PriviNet delivers advanced AI-driven IoT solutions, focusing on unbreakable connectivity and privacy-first intelligence. Its core technology, Lumra AI™, processes diverse data streams, including visuals and audio, to provide actionable, verifiable evidence from sensitive and remote environments. This transforms ambiguous alerts into trusted intelligence for applications ranging from in-home safety with Scout I to industrial asset monitoring with Scout X. Lumra AI enables low-power IoT devices to perform sophisticated analytics, optimizing resource use, reducing operational costs, and ensuring data integrity with advanced security protocols like encryption and blockchain. PriviNet's solutions are scalable and applicable across smart cities, precision agriculture, logistics, healthcare, airports, and environmental projects, driving innovation and improving quality of life.

Jobright is an AI-powered job search copilot designed to streamline and enhance the job application process. It offers personalized AI job matching, ensuring users see opportunities truly aligned with their skills and qualifications, free from fake listings. The platform features a 1-Click Application Autofill, allowing users to apply to numerous jobs daily across major ATS platforms, saving up to 80% of their time. Jobright also provides a job-specific tailored resume builder that crafts professional, ATS-compliant resumes in seconds. Additionally, it facilitates insider referrals by connecting users with alumni and hiring managers, potentially increasing interview chances by 4X. A 24/7 AI Career Copilot offers personalized guidance throughout the job search and interview preparation.

llama-dl offers a high-speed method for downloading Facebook's LLaMA 65B parameter GPT model, including all its various model weights (7B, 13B, 30B, 65B). This tool was created to provide a faster alternative to torrent downloads, achieving speeds of up to 40MB/s and completing the download in under two hours on a Chicago Ubuntu server. It utilizes a direct download link that was originally leaked, and the script has been updated to mirror the content to R2 after the original link was shut down. The project emphasizes accessibility for researchers and hackers to experiment with large language models, providing a convenient way to obtain the LLaMA model for personal use and research, while also addressing concerns about its licensing and potential commercial use.

LightGBM is a powerful and efficient open-source gradient boosting framework developed by Microsoft. It utilizes tree-based learning algorithms to deliver faster training speeds, higher efficiency, and lower memory usage compared to other boosting frameworks. LightGBM supports parallel, distributed, and GPU learning, making it capable of handling large-scale datasets effectively. It is widely used for various machine learning tasks including ranking, classification, and other predictive modeling. The framework is known for its accuracy and ability to achieve linear speed-up with multiple machines for training in specific settings. It provides extensive documentation, examples, and supports automated tuning through tools like FLAML and Optuna.

MAgent is a research platform specifically engineered for many-agent reinforcement learning, distinguishing itself from other platforms that typically focus on single or few-agent scenarios. It enables researchers to scale up their reinforcement learning experiments from hundreds to millions of agents, facilitating the study of artificial collective intelligence. The platform supports both Linux and OS X and allows for the implementation of various algorithms, including rule-based systems and deep learning frameworks. While the original project is no longer maintained, a community-maintained fork, MAgent2, is available for continued development and use. It offers examples for training and playing with agents in scenarios like pursuit, gathering, and battle, along with baseline algorithms like DQN, DRQN, and A2C.

llm-foundry is a comprehensive open-source repository offering code for the entire lifecycle of Large Language Models (LLMs), from training and finetuning to evaluation and deployment. It is specifically designed to integrate with Composer and the MosaicML platform, providing an efficient and flexible environment for rapid experimentation. The codebase supports various LLM workloads, including data preparation, training HuggingFace and MPT models from 125M to 70B parameters, and benchmarking training throughput and MFU. It also facilitates inference by converting models to HuggingFace or ONNX formats, generating responses, and evaluating LLMs on academic or custom in-context-learning tasks. The repository includes support for DBRX and MPT models, with detailed instructions for local use and community contributions.

Meeting Assistant Flow is an open-source project built on the crewAI framework, designed to streamline the entire meeting lifecycle. It automates critical tasks such as loading meeting notes from a text file, generating actionable tasks from meeting transcripts using AI agents, and integrating these tasks with Trello for project management. Additionally, it saves new tasks to a CSV file and sends Slack notifications to keep teams informed. This flow leverages multiple AI agents to handle different aspects of the meeting workflow, offering a modular and efficient solution for automating meeting management processes. Users can customize agents, tasks, and the flow itself to fit specific organizational needs.

Agent.so is an all-in-one platform designed for creating and utilizing AI agents to boost productivity and business growth. Users can chat with hundreds of pre-existing AI agents across various industries, or create their own custom AI agents from scratch without coding. The platform allows for training agents with specific data, files, and links, and equipping them with various skills. Agent.so emphasizes privacy with no data sharing and offers features like secure instant chat links and enterprise-ready solutions. It aims to provide a powerful AI agents network for 24/7 productivity, catering to individuals, freelancers, startups, and larger organizations.

mcp-context-forge is an open-source AI Gateway, registry, and proxy designed to federate Model Context Protocol (MCP) servers, A2A servers, and REST/gRPC APIs into a unified endpoint. It offers centralized governance, discovery, and observability across AI infrastructure, optimizing agent and tool calling. Key capabilities include a Tools Gateway for MCP, REST, and gRPC translation, an Agent Gateway for A2A protocol and OpenAI/Anthropic routing, and an API Gateway with rate limiting, authentication, and retries. The tool supports extensive plugin extensibility with over 40 integrations and provides OpenTelemetry tracing for comprehensive observability. It runs as a fully compliant MCP server, deployable via PyPI or Docker, and scales to multi-cluster Kubernetes environments with Redis-backed federation and caching.

mcp-server-chart is a Model Context Protocol (MCP) server designed for generating a wide array of charts using the AntV visualization library. This open-source tool supports over 25 different visual charts, making it suitable for various chart generation and data analysis tasks. It can be integrated with desktop applications like Claude, VSCode, and Cursor, or deployed via HTTP, SSE, or Streamable protocols for use with platforms like Aliyun and Dify. Key features include the ability to generate diverse chart types such as area, bar, boxplot, column, line, pie, scatter, and more, as well as specialized diagrams like fishbone, mind maps, and network graphs. Users can also filter available tools and configure private deployments for enhanced control over their chart generation services.

Mocha.jl is a deep learning framework for the Julia programming language, drawing inspiration from the C++ framework Caffe. Although now deprecated, it was designed for efficient training of deep and shallow convolutional neural networks, supporting optional unsupervised pre-training via stacked auto-encoders. The framework boasts a modular architecture with isolated components for layers, activation functions, solvers, and more, allowing for easy extension. Written in Julia, it offers a high-level interface for intuitive deep neural network experimentation. Mocha.jl provides multiple backends, including a portable pure Julia backend, a faster native extension backend, and a highly efficient GPU backend utilizing NVidia® cuDNN and CUDA kernels. It also supports HDF5 for data and model storage, ensuring compatibility with other computational tools, and can import Caffe model snapshots.

MockingBird is an open-source voice cloning tool designed for real-time speech generation. It allows users to clone a voice in approximately 5 seconds and generate arbitrary speech. The tool supports Chinese Mandarin and has been tested with multiple datasets, including aidatatang_200zh, magicdata, and aishell3. It is compatible with Windows, Linux, and even M1 macOS, offering flexibility for various environments. MockingBird leverages PyTorch and provides options for training custom models for encoders, synthesizers, and vocoders, or utilizing community-shared pretrained models. It offers a web server, a toolbox, and a command-line interface for generating voices.

MultiTalk is an innovative audio-driven multi-person conversational video generation framework, presented at NeurIPS 2025. It allows users to create videos featuring multiple characters engaging in conversations, singing, and other interactions, all driven by multi-stream audio input. Users provide a reference image and a prompt, and MultiTalk generates a video with consistent lip motions synchronized with the audio. Key features include support for both single and multi-person video generation, interactive character control via prompts, and generalization capabilities for cartoon characters and singing. The tool offers resolution flexibility (480p & 720p) and supports long video generation up to 15 seconds, with ongoing developments for longer durations and enhanced performance.

mlops-stacks offers a customizable, open-source solution for initiating new machine learning projects on Databricks, adhering to production best practices. It streamlines the development process by providing a pre-configured environment that includes ML project structure, ML resources as code, and CI/CD workflows (GitHub Actions or Azure DevOps). Data scientists can quickly iterate on ML code, while MLOps engineers can efficiently set up continuous integration and continuous deployment pipelines and manage ML resources. The tool supports automated model training and batch inference jobs across dev, staging, and production Databricks workspaces, facilitating an easy transition to production-grade ML solutions. It also integrates with Databricks asset bundles and offers options for Unity Catalog and Feature Store.

LOOKOUT is an advanced AI Marine Vision System engineered to significantly improve safety and situational awareness for boat operators. It addresses common causes of marine accidents, such as operator inattention and poor visibility, by integrating data from charts, AIS, computer vision, and radar into an intuitive 3D augmented reality view. The system features dedicated infrared night vision for confident navigation in darkness, wide-angle and zoom capabilities, and a 360º panorama view. This comprehensive solution provides an unparalleled safety advantage, allowing users to see through darkness and identify hazards often missed by traditional radar and AIS systems.

Noi, pronounced /nɔɪ/, is a browser extension focused on reducing chaos and increasing flow in browsing workflows. It provides an interaction-first approach for a tighter and more focused browsing loop. Key features include multi-window management for parallel workspaces, session isolation to keep contexts clean and predictable, and local-first data storage for history, prompts, and settings. Noi also offers prompt management for AI chats, allowing users to organize, reuse, and iterate on workflows. A built-in terminal for fast local commands and a CLI for controlling Noi from other tools like Claude Code and Gemini CLI further enhance its utility. Multiple themes and visual styles are available to fit different user setups.