AI Agents & Automation

Xtreme1 is an all-in-one open-source platform designed for multimodal training data, offering comprehensive solutions for data labeling, annotation, curation, and ontology management. It specifically supports 3D LiDAR point cloud, image, and Large Language Model (LLM) data, making it versatile for various machine learning challenges. The platform integrates AI-fueled tools to significantly boost annotation efficiency, supporting tasks like 2D/3D Object Detection, 2D/3D Semantic/Instance Segmentation, and LiDAR-Camera Fusion. Key features include built-in pre-labeling and interactive models, a configurable Ontology Center for managing classes and attributes, robust data management and quality monitoring, and tools for identifying and correcting labeling errors. Additionally, it provides model results visualization and a beta version for Reinforcement Learning from Human Feedback (RLHF) for LLMs.

Llongterm is an API designed to provide long-term memory capabilities for AI applications and agents. It allows AI chatbots and agents to remember past interactions and user-specific information, leading to more personalized and coherent conversations. The platform is compatible with various AI chatbots and agents, offering a human-readable memory format. Developers can integrate Llongterm to create AI teaching assistants, customer support bots, therapy note-takers, and product managers that maintain context across sessions. It supports both Javascript and Python, making it accessible for a wide range of AI development projects.

Dynamic Intelligence Asia (DIA) is an AI company based in Thailand, specializing in AI and computer vision solutions to enhance business efficiency, automation, and safety. Their offerings include a suite of AI-powered products like Smart AI Assistant DUSCAP, DIA Face Recognition, DIA Operator Monitoring, DIA Digital Meter Reading, and DIA Smart Analytic. DIA focuses on amplifying human capabilities through intelligent automation and computer vision, providing cost-saving solutions tailored for the Thai market. They emphasize customizable solutions, leveraging existing IP cameras as smart sensors, and process improvement through RPA and automated machine control. DIA prides itself on its own technology, R&D excellence, local team support, and deep understanding of local needs.

Futuria offers an advanced Agentic AI platform designed to transform industries by automating complex business processes and significantly boosting productivity. Their core product, Team™, is an adaptive work platform built to support real human teams in real-world, complex knowledge work environments. Unlike many AI solutions, Futuria's platform is engineered for high accuracy, repeatability, and dependability, integrating seamlessly with existing workforces and tech stacks. It goes beyond simple chatbots and automation, embedding true AI collaboration to create transformative efficiency for enterprises.

Abstra is an AI-powered finance automation platform designed to streamline financial operations, reduce manual work, and improve accuracy through intelligent workflows and integrations. It offers a development environment called Abstra Editor, available as a Web Editor for cloud-based development or a Local Editor for those preferring their own IDE and Git integration. The Abstra Cloud provides a robust environment to host, monitor, and manage production workflows, including features for project deployment, execution log monitoring, and team collaboration. Abstra leverages Python to build and automate business processes faster, making it suitable for organizations looking to enhance their financial efficiency.

awesome-llms-fine-tuning is a meticulously curated collection of resources designed to aid in the fine-tuning of Large Language Models (LLMs) such as GPT, BERT, and RoBERTa. This repository is an indispensable resource for machine learning practitioners and researchers looking to adapt pre-trained models to specific tasks and domains. It encompasses a wide range of materials including tutorials, academic papers, practical tools, robust frameworks, and established best practices. The collection helps users enhance model performance and ensure alignment with particular contexts, terminology, and application requirements. Whether expanding expertise or just starting in the LLM field, this repository offers valuable insights and guidelines to streamline the fine-tuning process.

Parallel AI is an end-to-end AI platform designed to automate and optimize business growth across the entire revenue journey. It offers smart lead generation through AI-powered prospecting and personalized outreach sequences, eliminating manual list building and data enrichment. The platform includes a Content Engine for creating high-quality, brand-aligned blog posts, social media content, and marketing copy, with direct publishing capabilities. Additionally, Parallel AI provides omni-channel AI Voice & Chat Agents that handle customer support, lead qualification, and appointment booking 24/7, integrating with existing CRMs and knowledge bases. A unique Executive Assistant feature proactively monitors business operations and delegates tasks, while white-label solutions allow agencies to offer these AI capabilities under their own brand.

Kodey AI transforms businesses by enabling the creation of autonomous AI agent teams without requiring any coding expertise. This platform allows users to build multi-agent systems where AI agents collaborate to automate complex workflows and scale operations. It focuses on delivering tangible results through intelligent AI orchestration, making advanced AI capabilities accessible to a broader audience. Kodey AI is designed for businesses looking to streamline processes, enhance efficiency, and achieve automation goals with agentic AI systems.

verl-agent is an extension of veRL specifically designed for training large language model (LLM) and vision-language model (VLM) agents using reinforcement learning (RL). It is the official code for the paper "Group-in-Group Policy Optimization for LLM Agent Training." A key differentiator is its step-independent multi-turn rollout mechanism, which allows for fully customizable per-step input structures, flexible history management, and modular memory. This design makes verl-agent highly scalable for very long-horizon, multi-turn RL training, such as tasks requiring up to 50 steps. The framework provides a diverse set of RL algorithms, including GiGPO, GRPO, and PPO, along with a rich suite of agent environments for both visual and text-based tasks, supporting models like Qwen3, LLaMA3.2, and LoRA fine-tuning.

Mindflow is a no-code, agentic automation and orchestration platform designed for Enterprise IT and Cyber professionals. It empowers teams to automate mundane tasks and deploy AI agents at scale, significantly enhancing productivity and operational agility. The platform features AI Agents that can reason and act across over 4,000 integrations, allowing users to build, deploy, and manage a 10x Digital Workforce. Mindflow supports use cases across ITOps, SecOps, CloudOps, and AIOps, offering capabilities like text-to-automation to turn prompts into ready-to-go flows and a catalog of 150,000 operations. It also includes secure on-premise access via a Relay Agent and AI-powered documentation generation for flows.

This repository offers PyTorch implementations of two significant transformer network architectures: "Attention is All You Need" by Vaswani et al. (NIPS 2017) and "Weighted Transformer Network for Machine Translation" by Ahmed et al. (Arxiv 2017). It serves as a valuable resource for researchers and developers interested in understanding, experimenting with, and applying transformer models in machine translation and other natural language processing tasks. The project is open-source, providing the full code for these implementations, making it accessible for academic study, practical development, and benchmarking against established models.

MindSpore is an open-source, self-developed AI framework by Huawei, designed to support full-scenario deep learning training and inference across device, edge, and cloud environments. It aims to provide a new AI programming paradigm, enabling developers to create efficient and flexible AI software and hardware applications. Key features include automatic differentiation based on source code transformation, automatic distributed parallel training, robust data processing capabilities, and a high-performance graph execution engine. MindSpore is primarily applied in AI fields such as computer vision and natural language processing, catering to data scientists and algorithm engineers. The framework fosters an AI development ecosystem through its open-source nature and community support, offering various toolkits for scientific computing, large models, and specific domains like CV and NLP.

Embedded LLM offers a no-code platform designed to help enterprises harness their data and build AI solutions rapidly. It simplifies the integration of Large Language Models (LLMs) into business workflows, allowing users to create powerful AI pipelines without coding. The platform features JamAI Base for collaborative AI and TokenVisor for managing and monetizing LLM workloads. Key capabilities include hardware acceleration optimized for AMD and multi-GPU clusters, robust data security with on-premise or private cloud deployment options, and full compatibility with the OpenAI API format. This ensures seamless migration and hybrid setups, preventing vendor lock-in and code rewrites. Embedded LLM aims to transform enterprise data into actionable AI solutions, boosting innovation and making advanced technology accessible to all business professionals.

Vibiscus is a deeptech company at the forefront of AI-powered noise control, specifically tackling the challenging problem of low-frequency noise (60-300Hz). Their flagship product, SilentFlow, is an active acoustic absorber engineered to target these frequencies within ducts. Unlike traditional passive or reactive silencers, SilentFlow is compact, repeatable, and easy to integrate into real-world HVAC deployments, offering a footprint 10 times smaller than existing systems. It ensures air circulation with minimal pressure loss and can reduce noise by up to -40dB. Vibiscus also provides acoustic diagnosis and customized solutions, which are fully deductible from a final SilentFlow order, ensuring tailored recommendations for noise-related issues.

Strivelabs is an experimentation and measurement engine designed for marketers, leveraging AI agents to streamline and enhance marketing efforts. It integrates with a wide array of existing tools such as Google Ads, HubSpot, Google Analytics, and Search Console, allowing AI agents to analyze data, generate ideas, and execute tasks across paid media, SEO, and content marketing. The platform emphasizes a 'human in the loop' approach, ensuring that all AI-driven recommendations and actions require explicit approval before going live. This allows marketing teams to run more experiments, measure what's truly working, and significantly reduce time spent on manual reporting and data analysis, ultimately accelerating pipeline growth.

Freeport Metrics is an American-owned and managed offshore software development partner specializing in generative AI integrations. They help startups and enterprises shorten the path to results with AI, offering services from product design and solution patterns to coding assistance and team training. Freeport Metrics builds and maintains AI systems for various operational problems, including voice agents, document intelligence, automation, and custom integrations. They also offer AI accelerators to ramp up product teams for the Gen AI age and provide solutions for leveraging existing applications with tools like Form2agent AI. Their expertise spans industries such as FinTech, InsureTech, HealthTech, Logistics, and FoodTech.

Asista provides AI Decision Intelligence (AIDI) solutions specifically designed for manufacturing operations. It leverages AI to transform complex data, operational constraints, and forecasts into prescriptive, execution-ready decisions. The platform helps organizations achieve margin recovery, generate executable revenue, and ensure profitable manufacturing by optimizing decisions across key enterprise functions. Asista addresses challenges like fragmented decision systems, inability to evaluate trade-offs, and static planning by offering solutions for production optimization, margin optimization, and prescriptive growth planning. It aims to deliver measurable impact through faster decision cycles and increased throughput and capacity.

EIP Solutions is a software development company dedicated to transforming innovative ideas into tangible digital solutions. They specialize in custom software development and cutting-edge applications, with a strong focus on emerging technologies like AI and machine learning. Their team of skilled professionals is committed to crafting solutions that not only meet but exceed client expectations, ensuring that developed products are current and future-proof. EIP Solutions serves a diverse range of industries, including healthcare, finance, restaurant, e-commerce, entertainment, social media, logistics, edtech, and real estate. They aim to be a partner in innovation, guiding clients through the process of turning aspirations into reality with breakthrough solutions.

Zephyr Global is an Artificial Intelligence studio dedicated to empowering businesses through digital transformation. Their flagship offering, Neo, acts as a 24/7 IT ally, providing expert assistance, strategic guidance, and extensive support across various platforms like Microsoft, AWS, Cisco, and Google Cloud, all at an affordable price. Beyond IT support, Zephyr Global offers innovative AI solutions such as AgriAssist, an AI-powered WhatsApp-based extension officer for farmers, and a 'Doctor in Your Pocket' service for instant medical second opinions. They also specialize in AI consulting and development, helping businesses automate processes like resume screening, marketing campaigns, and sales, ensuring real results and preparing them for the future of work.

AI-Infra-from-Zero-to-Hero is a comprehensive, open-source collection of resources dedicated to machine learning systems and AI infrastructure. It serves as a valuable guide for understanding the foundational and advanced concepts in this rapidly evolving field. The repository curates research papers, industry best practices, and video tutorials, covering critical areas such as Large Language Models (LLM) and Generative AI (GenAI). It includes insights from major conferences like OSDI, NSDI, SIGCOMM, SoCC, and MLSys, and references prominent models like Llama3 and Mistral. The project aims to provide a structured path for learning about AI system design, from data processing and training systems to inference and domain-specific infrastructures like AutoML and GNN systems.

AI-Functions offers an easy-to-use implementation for leveraging OpenAI's GPT-4 (or other model versions) to execute various tasks, significantly reducing the need for manual coding. Inspired by Ask Marvin, this project allows users to define function signatures, provide arguments, and describe the function's purpose, then let the AI generate the desired output. It's particularly useful for tasks like generating fake data or encrypting text. While powerful for many scenarios, the tool acknowledges limitations, especially with mathematical precision and complex calculations, where traditional algorithms might be more suitable. It includes test cases to evaluate performance across different GPT models.

AgentGuide is a comprehensive, open-source learning and interview guide for AI Agent development, designed to help users transition into large language model (LLM) roles. It offers systematic learning paths, practical projects, and job-oriented content, covering topics like LangGraph, advanced RAG techniques, and multi-agent systems. The guide is structured to take users from beginner to job-ready, with a focus on both algorithmic and development engineering paths. It includes a vast interview question bank, resume optimization tips, and real-world project examples to build a strong portfolio. AgentGuide emphasizes a "make something > learn something" philosophy, encouraging practical application and community engagement.

ELITE Research Lab LLC is dedicated to advancing artificial intelligence through focused research in key areas such as explainable AI, natural language processing, computer vision, and human-AI interaction. The lab's core mission is to deepen the understanding and practical application of AI technologies to address complex, real-world challenges ethically. They strive to develop AI systems that are not only powerful but also transparent and easily understandable, promoting trust and broader adoption. Their work aims to bridge the gap between theoretical AI advancements and their tangible impact on society, ensuring that AI solutions are both innovative and responsible.

DemoGPT is an open-source library designed to simplify the creation and deployment of LLM agents. It offers a comprehensive suite of tools, prompts, frameworks, and access to a knowledge hub of LLM models, all integrated into one platform. Developers can easily create custom tools and agents, leveraging built-in functionalities like TavilySearchTool, WeatherTool, and PythonTool. The library supports various agent types, including ToolCallingAgent and ReactAgent, which provides detailed reasoning processes for complex queries. DemoGPT also facilitates Retrieval Augmented Generation (RAG) with support for multiple vector stores, enabling agents to access and utilize document information effectively. This streamlines the entire agent development lifecycle, from prototyping to production.