AI Agents & Automation

ha-llmvision is a Home Assistant integration designed to bring visual intelligence to your home using multimodal large language models. It can analyze various visual inputs, including images, video files, live camera feeds, and Frigate events. The tool supports a wide range of AI providers such as OpenRouter, OpenAI, Anthropic, Google Gemini, AWS Bedrock, Azure, Groq, Ollama, Open WebUI, LocalAI, and any provider with OpenAI compatible endpoints. Key capabilities include answering questions and providing descriptions based on prompts, remembering people, pets, and objects, and maintaining a timeline of camera events. This timeline can be displayed on a dashboard or queried via Home Assistant's Assist feature, and sensors are seamlessly updated with extracted data.

Onigiri is an AI-powered language learning application designed to help users master English and Japanese by focusing on unknown vocabulary. Users can input text, and the tool intelligently identifies words they don't know, providing context-rich example sentences. It offers personalized example recommendations and unlimited audio streaming to enhance listening and speaking skills. The platform also includes progress tracking for individual words and provides access to over 20,000 original sentences with accompanying images and audio, all without requiring a login. This makes it a convenient and effective solution for self-directed language learners.

Rhetz is an AI running coach designed to transform your running experience through personalized guidance delivered directly via WhatsApp. This tool offers adaptive training plans that evolve with your progress, ensuring you always have an optimized regimen. It provides detailed performance analysis, helping you understand your strengths and areas for improvement. With Rhetz, runners can expect constant evolution in their training, making it suitable for anyone looking to enhance their running performance, whether for general fitness or specific goals like marathons. The convenience of WhatsApp integration makes coaching accessible and easy to follow.

The MultiPDF Chat App is a Python application built with Langchain that enables users to interact with multiple PDF documents. Users can ask questions in natural language, and the application will provide relevant responses by leveraging a language model. The process involves loading PDFs, extracting text, chunking the text, generating vector representations (embeddings) with a language model, and then matching user questions to semantically similar text chunks to generate answers. This tool is designed for educational purposes, serving as supporting material for a YouTube tutorial on building such an application. It requires an OpenAI API key for functionality.

LLMTools is an open-source Python library designed for efficiently running and finetuning Large Language Models (LLMs) in low-resource environments, specifically on consumer-grade GPUs. It features advanced finetuning capabilities in 2-bit, 3-bit, and 4-bit precision, leveraging the innovative ModuLoRA algorithm. The library provides an easy-to-use Python API for various tasks including quantization, inference, and finetuning. A key differentiator is its modular support for multiple LLMs, quantizers, and optimization algorithms, allowing for flexibility and integration with the HuggingFace Hub for sharing finetuned models. Developed as a research project at Cornell University, LLMTools is based on cutting-edge publications like ModuLoRA and QuIP, making it a valuable tool for researchers and developers working with LLMs.

LLM Sandbox is an open-source Python library designed to securely execute code generated by Large Language Models (LLMs) within an isolated environment. It offers a lightweight and portable sandbox runtime, ensuring safety through features like isolated execution, custom security policies, resource limits (CPU, memory, time), and network isolation. The tool supports various container backends, including Docker, Kubernetes, and Podman, and provides comprehensive language support for Python, JavaScript/Node.js, Java, C++, Go, and R. It seamlessly integrates with popular LLM frameworks like LangChain and LlamaIndex, and includes advanced features such as artifact extraction, on-the-fly library management, file operations, and container pooling for performance optimization.

The LLM-Engineers-Handbook is an official repository and practical guide for building end-to-end LLM-based systems, developed by Paul Iusztin and Maxime Labonne. It covers essential aspects from data collection and generation to LLM training pipelines, simple RAG systems, and production-ready AWS deployment. The handbook emphasizes LLMOps best practices, including comprehensive monitoring, testing, and evaluation frameworks. It details the use of various tools and cloud services like HuggingFace, Comet ML, Opik, ZenML, AWS, MongoDB, Qdrant, and GitHub Actions. The repository provides actively maintained code, installation instructions, and guidance on setting up local development and cloud deployment environments.

macai is a powerful and lightweight native macOS AI chat client designed to work with virtually any AI provider, including popular services like ChatGPT, Claude, xAI (Grok), Google Gemini, Perplexity, Ollama, and other OpenAI-compatible APIs. It boasts a user-friendly, minimalist light/dark UI and is packed with features such as vision capabilities, image generation, search, reasoning, and import/export options. A key differentiator is its iCloud Sync, which keeps chats, messages, and settings synchronized across all your macOS devices. The application prioritizes user privacy with no telemetry or usage tracking by macai itself, though Apple may collect anonymized telemetry if iCloud Sync is enabled. Users need to obtain API tokens for commercial LLMs, while open-source models can be run locally via Ollama.

MiniChain is a lightweight Python library designed for coding with large language models, offering a streamlined approach to prompt chaining. It enables developers to annotate Python functions for direct interaction with various language models and provides a visual graph of all calls for enhanced debugging and error handling. The library supports prompt engineering through Jinja templates, separating prompt text from code for better organization. MiniChain integrates with backends like OpenAI, Hugging Face, Google Search, and Python, and supports popular approaches such as Retrieval-Augmented QA, Chat with memory, and Chain-of-Thought. It also features a built-in visualization system using Gradio for interactive debugging and typed prompts for structured output generation.

Variational AI leverages advanced generative AI through its Enki™ platform to revolutionize early-stage drug discovery. Enki™ generates novel, synthesis-ready, lead-like compounds tailored to specific target product profiles, effectively eliminating the need for traditional hit identification and hit-to-lead phases. This allows biopharmaceutical partners to move directly into lead optimization with structures not discoverable by conventional methods. The platform optimizes across more than 50 parameters, including potency, selectivity, ADMET, and synthetic feasibility, for 760 pre-trained drug targets. By designing de novo molecular structures, Variational AI aims to provide better starting points for drug programs, leading to fewer costly iterations, faster timelines, and a higher probability of success.

Cognizen is a platform dedicated to leveraging artificial intelligence to enhance various industries and elevate different facets of life. The tool focuses on delivering cutting-edge AI solutions with a commitment to purpose, innovation, and integrity. It aims to create a future where AI acts as a powerful assistant, amplifying human potential rather than replacing it. While specific features are not detailed on the current website, the overarching goal is to provide advanced AI capabilities that drive progress and efficiency across diverse applications. Cognizen positions itself as a key player in the AI revolution, emphasizing responsible and impactful AI development.

DeepAudit is an open-source, multi-agent AI system designed to make code vulnerability detection and auditing accessible. It simulates the thought process of security experts through a collaborative architecture involving Orchestrator, Recon, Analysis, and Verification agents. This system aims to overcome common issues with traditional SAST tools, such as high false-positive rates, blind spots in business logic, and a lack of verification methods. Users can import projects, and DeepAudit will automatically identify tech stacks, analyze risks, generate scripts, perform sandbox verification, and produce professional audit reports. It supports Ollama for private deployment, ensuring data privacy, and has successfully identified numerous CVEs and GHSA security advisories.

MusPy is an open-source Python library designed to streamline the development of symbolic music generation systems. It offers a comprehensive suite of tools for various stages of the music generation pipeline, from data collection and preprocessing to model creation, training, and evaluation. Key features include a robust dataset management system with interfaces to PyTorch and TensorFlow, and extensive data I/O capabilities for common symbolic music formats like MIDI, MusicXML, and ABC. MusPy also provides implementations of various music representations, such as pitch-based, event-based, piano-roll, and note-based, catering to diverse generation approaches. Additionally, it includes model evaluation tools for audio rendering, score and piano-roll visualizations, and objective metrics, making it a valuable resource for researchers and developers in music AI.

Experts Vision Consulting (EVC) is a leading Saudi company providing SAP consulting and digital transformation solutions. They aim to contribute to Saudi Arabia's Vision 2030 by supporting the digital transformation journey of government and private sectors. EVC offers specialized expertise in areas such as strategic planning, beneficiary experience, innovation, governance, risk, and compliance, enterprise platforms and solutions, artificial intelligence, data governance, and smart cities. Their services include SAP implementation, support, and training, as well as community engagement programs like digital vision camps and innovation camps to foster technological skills and innovation.

QueryZy is an AI-powered data analysis tool designed for effortless interaction with various data files, including CSV, JSON, ARROW, and PARQUET. Users can upload their data and query it using natural language, eliminating the need for complex formulas or database knowledge. The tool offers instant data visualization through charts and allows for easy export of results. With a focus on privacy and security, QueryZy processes all data directly in the browser using DuckDB & WebAssembly, ensuring files never leave the user's device. It's ideal for quickly cleaning, cross-referencing, filtering, aggregating, and plotting data without any setup hassle or learning curve.

NVIDIA Model Optimizer is an open-source library designed to accelerate deep learning models through various state-of-the-art optimization techniques. It supports quantization, pruning, distillation, speculative decoding, and sparsity to compress models and enhance inference speed. The tool accepts Hugging Face, PyTorch, or ONNX models as input and provides Python APIs for composing optimization techniques. Optimized checkpoints can be seamlessly exported for deployment in frameworks like SGLang, TensorRT-LLM, TensorRT, and vLLM, making it a crucial component within the NVIDIA AI software ecosystem for efficient model deployment.

Fridge Leftovers AI is an innovative mobile application designed to help users reduce food waste and discover new meal ideas. By simply taking a photo of the contents of their fridge, the app leverages cutting-edge AI and image recognition technology to identify available ingredients. Users can then select the items they wish to use, and the AI generates a variety of mouthwatering recipes tailored to those specific ingredients. This intuitive process eliminates the need for manual ingredient entry, making meal planning effortless and inspiring creativity in the kitchen. It's ideal for anyone looking to maximize their ingredients and enjoy delicious meals without the hassle of traditional recipe searching.

Syntetica is an AI-powered platform designed to scale content production by enabling users to design, automate, and share content workflows without requiring any coding knowledge. It offers ready-to-use AI apps for various tasks, such as converting files into PowerPoint presentations, generating programming courses, or creating detailed sales contracts. Users can customize these solutions or build their own, incorporating different file types like text documents, spreadsheets, presentations, and images. The platform emphasizes ease of use, allowing teams to create and manage shared processes, automate routine tasks, and enhance productivity through instant automation and limitless creation. Syntetica also integrates with existing tools to streamline workflows and offers a welcome bonus of 10,000 tokens upon account creation.

OmniAI provides AI agents designed to put borrower onboarding on autopilot for lending businesses. These agents manage communication, collect necessary documents, and conduct follow-ups from start to finish, operating 24/7. The platform optimizes document workflows, allowing for conversational onboarding where borrowers can reply via SMS or email. OmniAI reads, verifies, and autofills fields, reducing friction and drop-off. It boasts 4x faster borrower onboarding, 98% confidence in compliance guardrails with 100% audit-trail coverage, and a 10-day shorter loan cycle. The system offers multilingual support and integrates with services like Plaid, Experian, LexisNexis, and TransUnion for unified data networking, enabling instant soft credit checks, bank account linking, income verification, and EIN retrieval.

gptme is a personal AI agent designed to operate directly within your terminal, offering a powerful suite of local tools. It can write and execute code, interact with the terminal, edit files, and browse the web, making it a versatile assistant for developers and knowledge workers. The tool supports various LLM providers, including Anthropic, OpenAI, Google, and local models via llama.cpp. A key feature is its extensibility through plugins, skills, and lessons, allowing users to tailor its functionality. gptme is also built for autonomous operation, enabling the creation of persistent agents that can run continuously, learn, and self-correct over time, with examples like 'Bob' managing tasks and interacting with platforms like GitHub and Twitter.

Groma is a cutting-edge grounded multimodal large language model (MLLM) developed by FoundationVision, designed with exceptional region understanding and visual grounding capabilities. It introduces a novel paradigm of grounded MLLMs by utilizing localized visual tokenization, distinguishing it from models that rely on LLMs for localization or external modules. Groma can process user-defined region inputs, such as bounding boxes, and generate detailed, long-form responses that are deeply grounded in the visual context. This makes it a powerful tool for developing multimodal assistants that require precise visual comprehension and contextualized output. The project is open-source, providing model weights and instructions for installation, data preparation, training, and inference, making it accessible for researchers and developers to customize and build upon.

LangChat is an open-source, Java-based AIGC project solution designed for enterprises. It integrates Role-Based Access Control (RBAC) and advanced AIGC large model capabilities, enabling businesses to rapidly build and customize AI knowledge bases and AI robot applications. The platform boasts extensive support for a wide array of AI large models, including DeepSeek, Gitee AI, Alibaba Cloud Tongyi, Baidu Qianfan, Douyin Doubao, Zhipu Qingyan, 01.AI, iFlytek Spark, OpenAI, Gemini, Ollama, Azure, and Claude. This multi-provider compatibility offers significant flexibility, allowing organizations to leverage the best-suited AI models for their specific needs within the Java ecosystem. LangChat aims to streamline the development of enterprise-grade AI solutions.

Lensmor is an Event Intelligence platform designed for B2B sales and GTM teams to optimize their trade show strategy. It decodes the "Who," "Where," and "Why" behind event attendees, providing verified emails and real-time budget signals to identify intent. The platform tracks over 160,000 global events, allowing users to search for high-value shows, unlock exhibitor directories, and access enriched decision-maker data including names, titles, verified emails, and LinkedIn profiles. Lensmor's AI Agent automates exhibitor discovery, scores targets by ICP fit, and executes personalized outreach, enabling teams to book meetings and generate pipeline before events even begin. It aims to transform trade show participation from relying on booth traffic to a proactive, data-driven approach.

Path Robotics offers intelligent welding cells powered by Obsidian AI, a foundational AI model specifically trained for welding. This technology addresses the skilled labor shortage in manufacturing by enabling real-time adaptation for high-quality welds, even with part-to-part variations. The system boasts 4x productivity, 30% lower costs, and a $0 CAPEX model, supported by 24/7 mission control. Obsidian AI learns from every weld, ensuring unmatched agility and continuous improvement, leading to exceptional weld quality with over 97% first-pass yield. Path Robotics serves industries such as manufacturing, critical infrastructure, defense, utility & energy, data centers, and heavy industry, providing solutions for fabricating structures like ship hulls, utility poles, and mining equipment.