The Future of Claude Code: Mods, Mutable Software, & Multiplayer Agents — Thariq Shihipar, Anthropic

Name: The Future of Claude Code: Mods, Mutable Software, & Multiplayer Agents — Thariq Shihipar, Anthropic
Uploaded: 2026-10-02T00:00:00.000Z
Duration: 5671 s
Channel: Latent Space
Description: Thariq Shihipar walks through Claude Code's next chapter: Mods that reshape the agent loop and UI, artifacts as a generative interface, multiplayer agents in Slack. He then recounts the OpenAI containment incidents candidly and explains why he backs external oversight of frontier labs.

LSLatent Space@LatentSpacePod

Full transcript

English

Summary：Thariq Shihipar walks through Claude Code's next chapter: Mods that reshape the agent loop and UI, artifacts as a generative interface, multiplayer agents in Slack. He then recounts the OpenAI containment incidents candidly and explains why he backs external oversight of frontier labs.

Watch on YouTube

Core points (3)

1Prompting is executive communication: front-load context; thin prompts waste your budget.
2Keep implementation notes: models consider the right answer, then talk themselves out of it.
3Match effort to domain; frontier models may beat small-model routing even on cheap tasks.