Don't just watch —
build the whole thing.
Ordered video courses that take you from nothing to a working system. Every episode is embedded right here, next to a written walkthrough and links to the deeper reference on each idea. Free, no sign-up.
Build a Real AI App.
A hands-on, build-along course: go from a naive 10-line RAG script that breaks at scale to a production retrieval-augmented app — chunking, embeddings and vector search, hybrid retrieval and reranking, grounded citations, the agent loop, evals, caching and streaming, prompt-injection defense, and deploy + monitoring. Every episode embeds the video and links to the deeper reference on each idea.
How LLMs Actually Work.
A visual, first-principles course on what's really happening inside a language model: attention, tokenization, embeddings, mixture-of-experts, fine-tuning and LoRA, RLHF, and diffusion. Seven short episodes, each embedded with a hands-on lab and the deeper reference — the mechanism, not the analogy.
The AI Engineer.
A concepts-first course on the craft of AI engineering: context and memory, embeddings and vector search, tools and agents, hallucination and grounding, the glue around the model, evals, deploy and monitoring, and prompt-injection security. Nine short episodes, each embedded with a hands-on companion and the deeper reference.
Data Structures & Algorithms.
The CS-interview core, made visual: hash tables, Big-O, binary search trees, heaps, tries, dynamic programming, and graphs. Seven short episodes, each embedded and paired with an interactive game in the Algorithms arcade — watch the idea, then play it.
Concurrency.
The topic that humbles confident engineers, made visual: race conditions, locks and mutexes, deadlock, threads vs processes, the async event loop, concurrency vs parallelism, and the actor model. Seven short episodes, each embedded with the deeper systems reference.
How the Internet Works.
Follow one web request from the moment you hit Enter to the pixels on your screen: DNS, IP and packets, TCP, HTTP, HTTPS/TLS, and CDNs. Seven short episodes that trace the whole journey end to end, each embedded with the deeper reference.
How Databases Work.
Open the black box: how indexes find one row in a billion, how transactions and the write-ahead log survive a crash, how isolation stops two writes from colliding, how caching absorbs read load, and how vector databases search by meaning. Five short episodes, each embedded with the deeper reference.
How Your Computer Works.
The layers under your code, one short episode each: the CPU, floating point, UTF-8, segmentation faults, garbage collection, flash memory, compression, public-key crypto, the speed of light, and what a container really is. Ten visual episodes that turn "it just works" into "I know why."
A course is the fast path in. The references are how deep you go.
Each course threads a single idea — a RAG app, the inside of a language model, the layers of the internet — into an ordered series of short episodes. You leave with a working system and a mental model, not a playlist you half-remember.
Build-along, not watch-along
Every course assembles one real system end to end. By the last episode you have the whole thing running — the point is the thing you built, not the hours you watched.
Three layers per episode
Each episode embeds the video, a written walkthrough you can skim or copy from, and links to the deeper handbook, lab, or system design on that exact idea. Watch it, read it, or go all the way down.
Ordered, but every episode stands alone
Go start to finish to build the whole system, or drop straight into the one episode you need today. Your progress is saved locally, so you can pick up exactly where you left off.
Free and self-contained
No paywall, no sign-up, no upsell. The videos, the write-ups, and every deeper reference they link to live here on the site, open to read.
Whether you're an engineer adding AI to your stack, a new grad building a portfolio, or switching careers into tech — start with a course, then follow the references as deep as you want to go.
Courses — frequently asked questions
Are the courses free?
Yes. Every course is free and self-contained — the videos, the written walkthroughs, and the links to deeper references. No paywall, no sign-up.
What is a Vibe Engines course?
A build-along video course: an ordered series of short explainers that together take you from nothing to a working system, with each episode embedded on the page next to a written summary and links to the deeper handbooks, labs, and system designs on each idea.
Do I need to watch the episodes in order?
They are designed to be watched in order — each episode builds on the app you assembled in the previous one — but every episode also stands alone if you only need one idea.
How long does a course take?
Most run around two hours of video across roughly a dozen short episodes, plus the reading and the code. You can complete one over a weekend or an episode at a time.