SUSHANT GUNDLA · LEARN← the site
← All courses

A hands-on course · 10 lessons

LLMs

The model layer, and how to choose in it.

What these models actually are, who sells them, what open weights really buy you, and how to pick one without guessing. The layer under everything else you build.

Lessons run about five to eight minutes each and end with a short recall quiz, so you find out what stuck before you move on.

Part 1The machine

01

What a model is made of

Frozen numbers, one repeated attention step, and two training stages. That is the whole product.

~7 min
02

Base, instruct, and reasoning models

Three different products wearing one word on the label. Picking the wrong one costs money and quality.

~7 min
03

Context windows and their limits

The window is a hard ceiling, and filling it is neither free nor harmless.

~7 min

Part 2The providers

04

The big three, and what actually differs

Anthropic, OpenAI, Google, and the layers you can buy them through — where the real differences live.

~7 min
05

What you actually pay for

Per-token billing, the three levers that move it, and how to see the bill before it arrives.

~7 min
06

Rate limits and reliability

Three separate ceilings, one status code, and a retry loop that does not make things worse.

~7 min

Part 3Open weights

07

Open weights vs closed APIs

What open really means here, what the licence actually lets you do, and the honest trade.

~7 min
08

Running a model yourself

What self-hosting actually takes — memory, quantisation, serving, and the arithmetic that decides it.

~8 min

Part 4Choosing

09

Picking a model

Start from your task and twenty of your own cases, not from a leaderboard.

~7 min
10

Switching models safely

Why a prompt does not transport, and the recipe for changing model without a bad week.

~7 min

Work through it in order the first time — each lesson leans on the one before it. After that, come back to any of them on their own.