Ask, Learn, Master.
https://arcgentic.smithgajjar.dev
Your personal AI learning ecosystem. An open-source, multi-agent platform that autonomously crafts highly personalized learning content to help you master any subject.
Arcgentic is a comprehensive AI-powered learning environment designed to transform how you study and explore new topics. By leveraging multi-agent orchestration, the platform builds an adaptive curriculum specifically for you.
- Ask anything — Start a learning session on any topic from scratch, or provide your own study materials (PDFs, URLs) for the agents to analyze.
- Learn interactively — Engage with a dedicated tutor agent that guides you step-by-step. The chat experience is enhanced with inline visualizations, dynamic diagrams, and interactive widgets.
- Master the material — Unlike standard chatbot interfaces, Arcgentic automatically generates a complete asset pack to help you master the material:
- Deep-dive explanations
- Interactive concept roadmaps
- Presentation slides
- Podcast scripts
- Spaced-repetition flashcards
Arcgentic acts as your personal curriculum factory. Once it understands what you are trying to learn, our multi-agent workflow autonomously researches and synthesizes a complete, structured study-pack:
- Deep-Dive Explanations: Long-form, beautifully formatted reading materials.
- Roadmaps: Interactive diagrams showing how topics interconnect.
- Audio Podcasts: Generated dialog-style podcasts exploring the subject matter.
- Slide Decks: Ready-to-read presentations summarizing key takeaways.
- Flashcard Sets: Spaced-repetition cards designed for active recall.
Drawing inspiration from systems like Claude's Artifacts, your Arcgentic Tutor doesn't just reply with simple text. During a session, the tutoring agent will actively generate and render living UI components directly in your chat stream. When explaining complex ideas, the agent can auto-generate:
- Custom SVG illustrations
- Dynamic data visualizations and charts
- Mermaid-based architectural diagrams and flowcharts
- Interactive learning widgets
Get Arcgentic up and running locally with just a few commands. This builds and launches the full platform (UI, APIs, and Database) via Docker.
Make sure you have installed on your machine:
- Node.js ≥ 18 & pnpm ≥ 10
- Python ≥ 3.11
- Go ≥ 1.21
- Docker Desktop (or another Docker engine)
-
Clone the repository
git clone https://github.com/smithg09/arcgentic cd arcgentic -
Start the platform One command installs dependencies, launches all services via Docker Compose, and runs migrations:
make start
The application is now running locally!
- Web UI: http://localhost:5173 (Open this in your browser)
- Agent API: http://localhost:5001
- User/GraphQL API: http://localhost:8080
Local development with hot reload? Use
make setup-localinstead, thenmake devto run services with live reloading.
A polyglot, multi-service architecture built for modularity and performance. Each layer is independently deployable and uses best-in-class tooling for its domain.
| Layer | Technology |
|---|---|
| Frontend | TypeScript, React 19, Vite, Tailwind CSS v4, shadcn/ui |
| State & Routing | TanStack Router, TanStack Query, Zustand |
| User Service | Go 1.21+, Echo, GraphQL (gqlgen), sqlc |
| Agent Service | Python 3.11+, Flask, LangGraph, LangChain |
| Web Server | Nginx |
| Database | PostgreSQL |
| Containerization | Docker, Docker Compose |
| Monorepo | Turborepo, pnpm workspaces |
Production-grade infrastructure on AWS with fully automated CI/CD. Zero static credentials — GitHub Actions authenticates via OIDC federation.
| Component | Technology |
|---|---|
| Containerization | Amazon ECR |
| Orchestration | Amazon EKS (Kubernetes) |
| Ingress / TLS | AWS ALB + ACM Certificate |
| Secrets | AWS Secrets Manager → External Secrets Operator |
| Database | Amazon RDS (PostgreSQL) |
| CI/CD | GitHub Actions |
| Landing Page | GitHub Pages (React build) |
| Documentation | Docsify (via GitHub Pages) |
For developers, contributors, and those looking to run services outside of Docker, please refer to our dedicated documentation guides:
- Getting Started / Local Dev — Detailed setup instructions, hot-reloading configurations, and troubleshooting tips.
- Agentic Harness — Deep dive into the Python multi-agent system, including the Supervisor/Worker architecture, UI tools, and state persistence.
- Architecture Overview — Deep dive into the monorepo structure, multi-agent LangGraph flows, and tech stack details.
- AWS Deployment — Deploy to Amazon EKS with CI/CD via GitHub Actions, ECR, Secrets Manager, and ALB ingress.
- Contributing — Pull request guidelines, code of conduct, and development standards.