ViewTube

Ever wondered what's really happening inside ChatGPT, Claude, or Gemini? In this video, we break down how a Large Language ...

I use Gemini Notebook to explore concepts for my labs. Disclaimer The content presented in these videos is intended solely for ...

The Planning Domain Definition Language (PDDL) is the standard and modern language for modelling and subsequently solving ...

How LLMs Turn the Internet Into Numbers: Tokenization, Sequences, and Embeddings The video explains how large language ...

... in modeling again um and the actual work without having to get hung up on I don't understand how this coding language works ...

Language Model and Sequence Generation text explained in a clear and practical way. Learn how Language Model and ...

In this video, we dive deep into the inner workings of Large Language Models (LLMs) and the mathematical process of next-token ...

Perforated is a data-efficiency layer for PyTorch that helps ML engineers build better models with less data. Through model ...

Welcome to Module 8. This is the deployment module — the one where everything we have built over the last seven modules ...

Perfect for students learning **Linear Programming, Operations Research, and Mathematical Modelling**.

Why does scaling improve language models—and why isn't model size enough? Follow one next-token prediction to see how ...

This is a project-based introduction to computational mechanics. There are five modules with exercises, homeworks, and final ...

The opening unit of the course: what a language model actually is and how it works - no math, minimal jargon. The core idea: a ...

In this video, I implemented a custom variant of GRPO from scratch, and used it to train Qwen 3.5 2B on the task of simulating the ...

What changes when a language model remembers two characters instead of one? We build a trigram model by hand, then ...

In this 4-minute walkthrough, I demonstrate how to supercharge your AI coding assistant in Visual Studio Code for Operations ...

The whole reward-model-plus-PPO pipeline collapses into a single classification loss on pairs of answers a human preferred ...

In this final video of the Swivel Embedding Series, we build a simple Swivel model from scratch using Keras. We start with the ...

In this video, we explore IBM Granite 4.2, the latest generation of IBM's open-source large language models built for enterprise AI, ...

DeepSeek-R1 is a large language model (LLM) designed to achieve elite reasoning capabilities through reinforcement learning ...