DeepTutor is a full-stack AI-powered learning companion that provides chat, problem-solving, research, visualization, and structured learning capabilities. The system features a modular architecture with a Python/FastAPI backend, Next.js/React frontend, and a CLI interface. It implements an agent-based system where capabilities like deep solve, deep question, and deep research are implemented as loop capabilities with access to tools such as web search, code execution, and knowledge base retrieval. The platform includes a three-layer memory system for contextual awareness, supports multiple RAG backends (LlamaIndex, PageIndex, GraphRAG, LightRAG, IMA), and enables integration with external knowledge bases (Obsidian, MarginNote 4) and partner agents (Claude Code, Codex). The application employs a plugin architecture for extensibility, allowing new capabilities, tools, and subagents to be added via entry points. Configuration is managed through JSON/YAML files with runtime overrides, and services are decoupled via provider abstractions for LLMs, embeddings, search, and media generation. Real-time communication is facilitated by WebSockets and an internal event bus (StreamBus). The system supports multi-user deployments with role-based access control, isolated workspaces, and grant-based resource sharing for skills, knowledge bases, and partners. Deployment options include local installation, Docker containers, and CLI-only mode. The codebase demonstrates strong engineering practices with extensive unit testing, clear separation of concerns, and reusable patterns such as the capability protocol, tool composition, and atomic file operations. While the system's feature richness introduces complexity, its modular design allows for selective adoption of components.
Llama Tutor is an open-source AI personal tutor built as a Next.js application. It allows users to enter a topic and select an education level (elementary to graduate), then uses Exa search to fetch relevant web content and Together AI (Qwen2.5-7B-Instruct-Turbo) to generate a personalized lesson in streaming markdown. The application features conversational follow-ups, source display, and rate limiting via Upstash Redis. It integrates Helicone for observability and Plausible for analytics. The codebase is a monolith with App Router, client-side state, and edge-deployed API routes. The architecture is straightforward: the main page orchestrates chat and source retrieval, relying on a custom streaming utility for Together AI responses. While the components are functional and well-structured for a demo, only the streaming utility stands out as a potentially reusable pattern for other AI-powered applications.