303 - How LLMs Work - the 20 minute explainer

Fragmented - The Software Podcast by Kaushik Gopal

Feb 2, 202600:25:45Technology

Ever get asked "how do LLMs work?" at a party and freeze? We walk through the full pipeline: tokenization, embeddings, inference — so you understand it well enough to explain it. Walk away with a mental model that you ca...

Related Episodes

Continue listening to more episodes from Fragmented - The Software Podcast.

310 - Mitchell Hashimoto on Ghostty & His Agentic Coding Workflow

Apr 14, 2026 - 00:59:43

Mitchell Hashimoto co-founded HashiCorp, built some of the most impressive DevOps tools like Vagrant and...

309 - Background Agents

Apr 1, 2026 - 00:25:37

Andrej Karpathy says the goal is to maximize how long an agent runs without your intervention. But there'...

308 - How Image Diffusion Models Work - the 20 minute explainer

Mar 24, 2026 - 00:24:44

You already know how LLMs work from our popular 20-minute explainer. Now we take it to images. What does...

307 - Harness Engineering - the hard part of AI coding

Mar 17, 2026 - 00:29:55

The hard part of AI coding isn't generating code — it's controlling quality, safety, and drift. Kaushik a...

306 - Keeping your agent instructions in sync and effective

Mar 10, 2026 - 00:23:22

AGENTS.md is becoming the common language for AI coding tools, but keeping repo rules, personal rules, an...

305 - Subagents explained - What they are, when (not) to spawn them

Feb 17, 2026 - 00:26:38

Subagents are becoming a core primitive for serious AI-assisted development. In this episode, Kaushik and...

304 - Agent Skills - when to use them and why they matter

Feb 9, 2026 - 00:26:47

Agent Skills look simple, but they are one of the most powerful building blocks in modern AI coding workf...

302 - MCPs Explained - what they are and when to use them

Jan 26, 2026 - 00:19:09

MCPs are everywhere, but are they worth the token cost? We break down what Model Context Protocol actuall...