AutoGen

Top Builders

Explore the top contributors showcasing the highest number of app submissions within our community.

AutoGen

AutoGen is an advanced open-source framework developed by Chi Wang designed to simplify the creation of multi-agent systems powered by large language models (LLMs). The platform allows developers to build conversational AI agents that can interact with each other, humans, and various tools in a coordinated manner. AutoGen is highly modular and supports a wide range of applications, making it an essential tool for developers looking to implement complex, automated workflows with minimal manual intervention.

General
Author	Chi Wang
Release Date	September 2023
Website	https://microsoft.github.io/autogen/
Repository	https://github.com/microsoft/autogen
Documentation	https://microsoft.github.io/autogen/docs/Getting-Started
Discord	https://discord.com/invite/pAbnFJrkgZ
Technology Type	AI/ML Framework

Key Features

Multi-Agent Framework: Facilitates the design of agents with specialized roles, enabling them to communicate and collaborate efficiently.
Enhanced LLM Inference: Provides advanced APIs for improving LLM performance, reducing inference costs.
Customizable Workflows: Supports complex, dynamic workflows by allowing agents to interact through conversational patterns, enabling seamless automation.
Tool Integration: Agents can be configured to use external tools, adding flexibility and enhancing their problem-solving capabilities.
Human-in-the-Loop: Integrates human feedback into the workflow, allowing for oversight and intervention when necessary.

Start Building with AutoGen

AutoGen simplifies the development of complex AI applications by providing a robust framework for creating multi-agent systems. With its modular design, developers can quickly build and customize AI workflows that combine LLMs, human intelligence, and various tools to tackle intricate tasks. Whether you are looking to automate customer support, enhance software development processes, or optimize supply chains, AutoGen offers the flexibility and power needed to create sophisticated AI-driven solutions. Explore the community-built use cases and applications to see the full potential of what AutoGen can do.

👉 Start building with AutoGen

👉 Examples

Edit on GitHub

AutoGen AI technology page Hackathon projects

Discover innovative solutions crafted with AutoGen AI technology page, developed by our community members during our engaging hackathons.

InvestiGuard AI

InvestiGuard AI is a multi-agent fraud investigation platform built for regulated and high-stakes financial workflows. Traditional fraud detection systems often provide risk scores with limited transparency, making investigations difficult to audit and explain. My solution uses Band to coordinate five specialized AI agents that collaborate throughout an investigation. The Intake Agent triages suspicious cases, the Transaction Analyst examines transaction patterns, the Behaviour Analyst reviews customer activity, the Risk Assessor evaluates evidence and calculates risk, and the Lead Investigator generates the final recommendation. Rather than relying on a single AI system, agents exchange messages, delegate tasks, share findings, and build upon each other’s analysis through a structured collaboration workflow. The platform visualizes this process through an Agent Collaboration Network, Band activity feed, evidence board, investigation timeline, and collaboration metrics such as handoffs, messages exchanged, active agents, and confidence scores. To support accountability, InvestiGuard AI includes a human-in-the-loop review stage where investigators can inspect evidence, review recommendations, approve or override decisions, and generate a complete audit report. The demo showcases multiple fraud scenarios including transaction structuring, suspicious international wire transfers, and legitimate high-value transactions. By combining multi-agent collaboration, explainable reasoning, human oversight, and auditable decision-making, InvestiGuard AI demonstrates how Band can power trustworthy AI systems for fraud detection, risk assessment, and financial compliance workflows.

Oner: The Continuous-Shot Developer Swarm

Oner is an autonomous, cross-framework developer swarm packaged as a full-stack Next.js web application. Designed to execute software engineering from planning to visual QA in one seamless workflow, it provides users with a premium, split-screen GUI to monitor AI agents in real time. Our absolute standout feature and the core of our business value is the Dynamic Model Routing engine. Traditional multi-agent systems burn through expensive API credits blindly. Oner solves this natively. When a user submits a prompt, our LangChain-based dispatcher, The Librarian, leverages the high-tier reasoning of the AI/ML API to analyze the project's complexity. It then dynamically selects and routes the most efficient LLMs to the rest of the swarm based on task requirements. This allows us to seamlessly swap between cost-saving models for brute-force execution and premium models for complex reasoning, proving a highly scalable architecture. This dynamic routing dictates a roster of 6 specialized agents built across diverse frameworks (LangChain, LlamaIndex, AutoGen, and Native Node.js). To prevent context window bloat, these disparate frameworks are united entirely by the Band collaboration layer, which acts as the definitive shared state and communication bus for all handoffs. Finally, Oner crowns its pipeline with Glassion, a multimodal Anti-Slop QA agent that evaluates headless browser screenshots against high-taste design principles to enforce perfect spatial harmony. Combined with a Band-powered Human-in-the-Loop (HitL) escalation protocol, Oner is the ultimate enterprise-ready builder.

OmniSentinel: Watching Everything, Missing Nothing

OmniSentinel is an autonomous multi-agent crisis intelligence platform that helps organizations, governments, and emergency teams detect, understand, and respond to critical events in real time. Modern crises such as natural disasters, cyberattacks, infrastructure failures, and public health emergencies generate vast amounts of fragmented information that can overwhelm human decision-makers. OmniSentinel transforms this complexity into coordinated intelligence through a Band of AI Agents working together as a unified command center. The platform combines specialized agents that continuously monitor data, identify threats, gather intelligence, forecast potential developments, simulate response scenarios, and recommend optimal actions. Instead of operating independently, these agents collaborate, validate findings, and refine recommendations to provide accurate and actionable insights. This collective reasoning approach enables faster situational awareness and more confident decision-making during high-pressure events. OmniSentinel provides a real-time view of evolving situations, helping users understand current risks, anticipate future impacts, and evaluate possible response strategies before taking action. For demonstration purposes, the platform can simulate scenarios such as cyberattacks, severe weather events, infrastructure outages, or large-scale emergencies, showcasing how multiple AI agents coordinate autonomously to analyze information and generate strategic response plans within seconds. By turning information overload into coordinated intelligence, OmniSentinel empowers decision-makers to respond faster, smarter, and more effectively when every second matters.

Human First Capital

Finding the right people for a hackathon—whether teammates, organizers, or future hosts—is often a manual and fragmented process spread across dozens of websites and communities. This project uses BrightData's MCP to create an autonomous talent discovery agent. Instead of repeatedly searching the web myself, I provide a configuration describing the people I'm looking for, and the agent continuously gathers and evaluates candidates from across the internet. Beyond discovery, the system creates rich profiles of individuals using publicly available information. Rather than simply collecting links and social accounts, it builds a narrative around a person's work, projects, interests, collaborators, and contributions over time. This allows users to understand not only what someone has built, but also who they have worked with, what communities they participate in, and how their interests have evolved. The system learns from my feedback, improving future recommendations and surfacing people who match my interests, experience, and past hackathon projects. It can also identify past hackathon organizers and people planning future events, helping users build stronger connections within the hackathon ecosystem. Using this system, I have already discovered several impressive builders and multiple hackathons—both past and upcoming—that I likely would not have found through traditional search alone. What makes this project stand out is that it transforms web discovery from a one-time search into a continuously running, feedback-driven process. By combining autonomous web discovery, profile synthesis, relationship mapping, and personalized learning, the system acts as a persistent research partner that helps users discover opportunities and the stories behind the people creating them. The result is a continuously improving discovery engine that turns web research into an automated workflow.

LexAfrica AI

Your description is 139 characters over the limit! Let me trim it to exactly fit: 🌍 THE PROBLEM: 1.4 billion Africans lack access to legal services. Legal consultations cost $100-500 - unaffordable for 85% of the population, leaving millions without legal recourse. ⚖️ OUR SOLUTION: LexAfrica AI employs 4 specialized AI agents powered by Bright Data's live web intelligence: • Intake Agent: Classifies legal domains and extracts key facts • Research Agent: Fetches real-time legal information via Bright Data SERP API and Web Unlocker • Advisor Agent: Provides plain-language legal advice with actionable steps • Document Agent: Drafts formal legal letters citing relevant laws 🌐 BRIGHT DATA INTEGRATION: Our key differentiator is live web intelligence. Instead of relying on outdated AI training data, we use Bright Data's SERP API to fetch current legal information from authoritative sources. Users see verifiable citations with clickable URLs, ensuring transparency and accuracy. ✨ KEY FEATURES: • Multi-language support (6 African languages: English, Swahili, Hausa, Yoruba, French, Amharic) • Voice input for low-literacy users • Mobile-responsive design with light/dark mode toggle • Real-time agent visualization • PDF document generation • Live web sources with clickable verification links 🚀 IMPACT: We've transformed legal access from $100-500 consultations taking weeks to free, 30-second AI assistance. Our solution addresses tenant rights, employment disputes, contract issues, and family law. 💡 TECHNICAL ARCHITECTURE: Built with Llama 3.3 70B via Groq, FastAPI backend on Railway, Next.js frontend on Vercel, and Bright Data APIs for live web intelligence. This isn't just an AI demo - it's a production-ready solution that could help millions access justice.