DeepTutor vs How LLMs Work

DeepTutor is an open-source, agent-native personalized learning platform from HKU's Data Intelligence Lab. Unlike chatbot-style tutors, it introduces "TutorBots" — persistent autonomous agents assigned to a specific subject or course, each with their own workspace, memory, and context. You don't start over every session; the TutorBot knows where you left off and what you're struggling with. The platform ships five unified learning modes — Chat, Deep Solve, Quiz Generation, Deep Research, and Math Animator — all sharing context through the TutorBot memory layer. Deep Solve breaks problems into sub-tasks, runs web searches and code execution, and builds up explanations step by step. Math Animator renders LaTeX expressions as Manim animations. Under the hood it supports 28+ LLM providers (Anthropic, OpenAI, Ollama, local models), full RAG on uploaded documents, and a CLI plus Docker support for self-hosting. Version 1.0.0 shipped in April 2026 after hitting 10,000 stars in 39 days earlier in the year. It's one of the few open-source AI education projects that treats the learner as a long-term relationship rather than a one-off query. This is the architecture that matters for AI in education — not tutors that forget you.

"How LLMs Work" is a free, browser-based interactive guide that walks through the complete pipeline for building large language models — from raw web scraping to RLHF-trained conversational assistant. Created by Yash Narwal and based on Andrej Karpathy's technical deep-dive lecture, it's been getting significant traction on Hacker News (214+ points) for turning dense ML theory into something genuinely accessible. The site covers data collection and deduplication, Byte Pair Encoding tokenization with a live demo, pre-training and next-token prediction, inference with a probability sampling simulator, post-training with RLHF, and RAG. Each section uses animated visualizations, clickable pipeline diagrams, and canvas-based graphics — not static explainer images. The progressive narrative structure follows a single piece of text through every stage of the pipeline, making abstract concepts concrete. In an era where everyone uses LLMs but few understand how they work, this kind of high-quality educational resource matters for a different reason than tools: it raises the baseline competency of the entire developer ecosystem. Better-informed builders ask better questions, make better design decisions, and push vendors toward more transparency. This is the kind of project the HN community rewards — and deserves the signal boost.