310 - Mitchell Hashimoto on Ghostty & His Agentic Coding Workflow

More from Fragmented - AI Developer Podcast

1. 

   E311 · 6 Sep 2026 · 13 min

   311 - Self learning harnesses are here

   Claude Code and Codex have been the mainstay. OpenCode and Pi showed what open-source harnesses could be. But two new entries just moved the goalposts: DeepSeek's Cordis kernel treats everything — even the agent loop itself — as a hot-swappable plugin, and Prime Agent runs the whole harness inside a live REPL where sub-agents are just function calls. Self-learning harnesses are here, and they're open source.

2. 

   E309 · 1 Apr 2026 · 26 min

   309 - Background Agents

   Andrej Karpathy says the goal is to maximize how long an agent runs without your intervention. But there's a false summit most teams hit first: individual speed goes up while system speed stalls, your laptop roars under four parallel Gradle builds, and review queues back up. Kaushik and Iury trace the full arc — from local multitasking to cloud-hosted async work to fully autonomous agents that fire on repo events and put PRs in your inbox. Full shownotes at https://fragmentedpodcast.com/episodes/309

3. 

   E308 · 24 Mar 2026 · 25 min

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

   You already know how LLMs work from our popular 20-minute explainer. Now we take it to images. What does Michelangelo have to do with stable diffusion? More than you'd think. Walk away knowing how image generation actually works — and what it has in common with the text models you already understand. Full shownotes at https://fragmentedpodcast.com/episodes/308/

4. 

   E307 · 17 Mar 2026 · 30 min

   307 - Harness Engineering - the hard part of AI coding

   The hard part of AI coding isn't generating code — it's controlling quality, safety, and drift. Kaushik and Iury break down harness engineering: the five pillars for shaping an agent's environment and what it looks like when teams build custom harnesses from scratch. Full shownotes at - https://fragmentedpodcast.com/episodes/307

5. 

   E306 · 10 Mar 2026 · 23 min

   306 - Keeping your agent instructions in sync and effective

   AGENTS.md is becoming the common language for AI coding tools, but keeping repo rules, personal rules, and tool-specific files in sync is still messy. In this episode, Kaushik and Iury break down the sync problem, compare their own setups, and unpack what the latest AGENTS.md research actually says. Full shownotes at https://fragmentedpodcast.com/episodes/306/

6. 

   E305 · 17 Feb 2026 · 27 min

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

   Subagents are becoming a core primitive for serious AI-assisted development. In this episode, Kaushik and Iury disambiguate "agent" terminology, unpack plan mode vs subagents, and explain how parallel, scoped workers improve research quality without polluting the main thread. Full shownotes at https://fragmentedpodcast.com/episodes/305

7. 

   E304 · 9 Feb 2026 · 27 min

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

   Agent Skills look simple, but they are one of the most powerful building blocks in modern AI coding workflows. In this episode, Kaushik and Iury break down when to use skills, how progressive disclosure works, and how skills compare with commands, instructions, and MCPs. Full shownotes at https://fragmentedpodcast.com/episodes/304

8. 

   E303 · 2 Feb 2026 · 26 min

   303 - How LLMs Work - the 20 minute explainer

   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 can use for your next dinner party.

9. 

   E302 · 26 Jan 2026 · 19 min

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

   MCPs are everywhere, but are they worth the token cost? We break down what Model Context Protocol actually is, how it differs from just using CLIs, the tradeoffs you should know about, and when MCPs actually make sense for your workflow. Full shownotes at https://fragmentedpodcast.com/episodes/302

10. 

    E301 · 19 Jan 2026 · 25 min

    301 - The AI coding ladder

    Most folks reference "AI coding" like it's one thing. It's really not. In this foundational episode Kaushik & Iury walk through (at least) four paradigms — from super autocomplete to agent orchestration — each with different workflows, expectations, and mental models. What do most developers follow today? Where is the frontier? What's coming in the future? Listen to the episode and find out!

Every episode of Fragmented - AI Developer Podcast →