DigitableCourses
Portal settings
Русский
Show me around

Local and account-free. No account, no sign-up: whatever you type into the tools stays in this browser's localStorage and never reaches a server.

Our own counter records page opens and finishes: the address leaves, plus one tenth-of-the-text figure. No cookies, no outside trackers, your IP is not stored, Do Not Track is honoured. How to check that

The portal's own repository is not published, so we do not call it open source. What is open:

Digit flang Agent gate tools-core IT Tools Chat digitwm blender-mcp Hugging Face Licences

The portal lives on donations, paid consultations and requested write-ups, and on Workbench sales.

Support Order a write-up Workbench

There are no plans to make the courses paid.

About this video

In this video:

  • how LLMs, neural networks, and machine learning algorithms based on SBSE work
  • why neural networks don’t “think”
  • what tokens and context windows are
  • how professional prompting works: ZS, FS, RCTF, ToT, CoT
  • what RAG, MCP, and AI agents are
  • why graphics cards are needed
  • how images are generated
  • and why AI is no longer “magic” but an engineering tool

The video will be suitable:

  • for developers
  • for those who are just starting to learn about AI
  • and for everyone who wants to understand how it works without hype and marketing

📌 The video is based on a stream — without cutting out context and with explanations along the way.

Text version

The lecture has been broken down into a text track of 16 chapters — AI for Engineers: Fundamentals. It can be read without a player, searched for on the portal, and each chapter links to the relevant minute of this recording.

Chapter Timestamp
Predictability: what question does AI actually solve? 00:00
Markov chains: prediction without understanding 05:00
Machine learning: images, sound, text 12:15
Neural networks: how to train a model 19:32
Representations: why the model “doesn’t know” what the world looks like 21:10
Large language models: from T9 to transformer 30:13
Tokens, context window, and dialogue memory 38:04
Hardware: why AI needs graphics cards 36:39
Can neural networks think? 34:20
Prompting: RCTF and STAR 43:32
Advanced prompting: ZS, FS, CoT, CoV, ToT 53:53
Diffusion models: how images are generated 01:00:30
RAG, MCP, and agents 01:14:51
Providers, framing, and censorship 01:20:55
SDD, orchestration, and discussions about replacing specialists 01:35:30

Timestamps

00:00 — Научный вопрос и предсказуемость
05:00 — Цепи Маркова
11:12 — Основные вехи ИИ
11:30 — SBSE
12:15 — Машинное обучение: картинки
15:36 — Машинное обучение: аудио
17:22 — Машинное обучение: текст
19:32 — Нейросети и обучение моделей
21:10 — Мы не знаем как выглядит мир
23:38 — Поиск взаимосвязей
30:13 — Большие языковые модели
32:15 — Как LLM понимает текст
34:20 — Могут ли нейросети думать
36:39 — Зачем нужны видеокарты
38:04 — Токеномика
43:32 — Промптинг
47:10 — STAR
53:23 — Польза промптинга
53:53 — Advanced Prompting
53:56 — Zero-shot prompting
54:54 — Few-shot prompting
56:06 — Chain of Thoughts
57:01 — Контекстное окно у LLM
58:02 — Как LLM помнит ваши запросы
01:00:30 — Диффузионные модели
01:01:58 — Как генерируются картинки
01:03:02 — Мотивация провайдеров на оптимизацию
01:06:05 — Chain of Verification
01:06:14 — Tree of Thoughts
01:09:00 — RCTF
01:14:51 — RAG
01:16:00 — Кому нужен промптинг?
01:16:29 — Кому ИИ может помочь?
01:20:55 — DeepSeek vs ChatGPT vs Gigachat
01:21:45 — Фрейминг и цензура
01:35:30 — SDD (Spec-Driven Development)
01:40:34 — Оркестрация
01:42:56 — Подводим итоги
01:46:57 — Самый важный подход для применения ИИ сегодня
01:49:49 — Коротко про MCP и skills
01:51:40 — Про замену специалистов на ИИ