Documentation for AI coding tools, turned into a lecture you can listen to on the way to work.
This is the STARTING track: it begins at zero and takes you to the point where you hand a task to an agent yourself and get back what you meant. The archive is a syllabus rather than a news round-up, so the episodes are meant to be heard in order.
Every episode says the date of the material it was written from, because these tools change every week and the date is part of the fact.
The episodes are written from the vendors' own documentation and are the author's own explanation of it — never a reading of it. Sources are named in every episode.
Why follow: new episodes land in your app on their own, with nothing to check — and the app keeps them in order, which matters in a course.
Every episode description lists the sources it stands on, the date the documentation was current, and the date it was last re-checked. These tools change weekly, so an episode without a date cannot be verified.
Daily news has its own show: “AI Toolkit: Pulse”.
An unofficial, independent podcast. Not affiliated with Anthropic, OpenAI, GitHub or Microsoft. Narrated with synthetic speech.
From nothing to a working agent in fifteen minutes — and the one thing to understand before you start is which directory you launch it in.
Sources:
· https://code.claude.com/docs/en/quickstart.md
· https://code.claude.com/docs/en/setup.md
Documentation as of 2026-08-03 · Last re-checked against sources: 2026-08-06
An unofficial, independent podcast. Not affiliated with Anthropic, OpenAI, GitHub or Microsoft. Narrated with synthetic speech.
6 Aug 2026
Plans, Prices, Limits: What You're Really Paying For
Subscriptions and limits change every quarter, but the logic is constant: you pay for model power and for how hard you use it — and you can start sensibly for very little money.
Sources:
· https://www.anthropic.com/pricing
· https://code.claude.com/docs/en/costs.md
Documentation as of 2026-08-03 · Last re-checked against sources: 2026-08-06
An unofficial, independent podcast. Not affiliated with Anthropic, OpenAI, GitHub or Microsoft. Narrated with synthetic speech.
5 Aug 2026
Chat vs. Agent: Why Claude Code Isn't a Chatbot in a Terminal
An agent differs from a chat in one thing: it has hands — it reads files, runs commands and checks the results of its own work, in a loop, until it is done.
Sources:
· https://code.claude.com/docs/en/overview.md
· https://code.claude.com/docs/en/agent-sdk/agent-loop.md
Documentation as of 2026-08-02 · Last re-checked against sources: 2026-08-06
An unofficial, independent podcast. Not affiliated with Anthropic, OpenAI, GitHub or Microsoft. Narrated with synthetic speech.
5 Aug 2026
Three Companies, Three Ways to Work: Anthropic, OpenAI, and Microsoft
You are not picking the best AI, you are picking a place to work — and these three differ in philosophy far more than in model quality.
Sources:
· https://code.claude.com/docs/en/overview.md
· https://platform.claude.com/docs/en/about-claude/models/overview.md
· https://docs.github.com/en/copilot/get-started/what-is-github-copilot
· https://platform.openai.com/docs
Documentation as of 2026-08-02 · Last re-checked against sources: 2026-08-06
An unofficial, independent podcast. Not affiliated with Anthropic, OpenAI, GitHub or Microsoft. Narrated with synthetic speech.
5 Aug 2026
The Claude Grid: How Haiku, Sonnet, Opus and the Rest Fit Together
Model families are tiers of power and price, version numbers are generations — understand that grid and you understand 80% of the AI news you read.
Sources:
· https://platform.claude.com/docs/en/about-claude/models/overview.md
· https://platform.claude.com/docs/en/about-claude/models/choosing-a-model.md
· https://raw.githubusercontent.com/anthropics/claude-code/main/CHANGELOG.md
Documentation as of 2026-08-01 · Last re-checked against sources: 2026-08-06
An unofficial, independent podcast. Not affiliated with Anthropic, OpenAI, GitHub or Microsoft. Narrated with synthetic speech.
4 Aug 2026
The Next-Token Machine: What the AI That Writes Code Actually Is
A language model is a next-token prediction machine, and everything follows from that one property: the power, the hallucinations, and the way you have to talk to it.
Sources:
· https://platform.claude.com/docs/en/about-claude/models/overview.md
· https://code.claude.com/docs/en/glossary.md
Documentation as of 2026-08-01 · Last re-checked against sources: 2026-08-05
An unofficial, independent podcast. Not affiliated with Anthropic, OpenAI, GitHub or Microsoft. Narrated with synthetic speech.
Contact AI Toolkit: Start
Guest appearances
Does not typically book guests
Based on episode analysis; this does not confirm that the show is currently accepting guests.
Host of AI Toolkit: Start?
Claim your podcast to manage its listing and keep your show details accurate.
Pod Engine is an independent podcast discovery and analytics service and is not affiliated with or endorsed by this podcast. Artwork and show content belong to their owners. Full legal notice.
Explore this show Podcast research with Pod Engine
