CtrlK
BlogDocsLog inGet started
Tessl Logo
PluginPublic

ainativedev/aidevcon-2026-nyc

AI Native DevCon 2026 New York — practical agent skills grounded in conference talks

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

Overview
Quality
Evals
Security
Files

outline.mdskills/talk-yegge-ai-powered-factories/references/

Source outline

Source: Steve Yegge, The Future of AI-Powered Factories, AI Native DevCon New York 2026. Recording. All anchors refer to original numbered cues in the transcript. Short quotations below preserve the supplied transcription, including spoken wording.

TopicSource anchorExact short quoteApplication and scope
Pets and reproducibility00:00:44–00:01:37, cues 13–28“it's totally non reproducible”The pets-versus-cattle analogy motivates reproducibility.
Execution layer00:04:01–00:06:26, cues 64–104“curation of the context”Reusable factory machinery differs from personal creative work.
Model-task fit00:06:30–00:07:34, cues 105–124“evaluate whether models are smart enough”The speaker describes evaluating task suitability, not training each model.
Bespoke entanglement00:07:53–00:09:49, cues 130–163“I was hurting the system by trying to open source.”The personal factory had contextual dependencies.
Overwhelming recipients00:09:37–00:10:14, cues 160–169“overwhelm your customers and your business colleagues”External capacity can constrain factory output.
Long-lived roles00:14:04–00:16:08, cues 236–275“They're iterative. They're collaborative.”The entourage layer is described as long-lived collaboration.
Accidental policy00:18:31–00:19:58, cues 323–350“they create case law.”Casual remarks could become durable rules and block work.
Memory and resources00:20:25–00:22:09, cues 358–386“any memory system has to have forgetting built into it”The speaker reports memory accumulation and neglected operational limits.
Creative judgment00:23:27–00:24:48, cues 406–428“go make a bunch of candidates.”Iterative comparison helps express human taste.
Business pace Q&A00:31:34–00:33:32, cues 534–566“move slow, go at human speeds,”The advice emphasizes value and absorption capacity.
Attention and economics Q&A00:33:39–00:36:19, cues 567–615“They did all the easy stuff. It's exhausting.”Personal experience motivates attention and cost constraints.
Feedback bias Q&A00:36:24–00:37:37, cues 616–636“they're not doing anything that helps me with a new player experience.”Existing-player feedback can overlook newcomers.

Source limitations

  • Wheelhouse and game outcomes are personal reported experience, not a transferable reference implementation or controlled benchmark.
  • Model names, pricing, subscription arrangements, and future cost predictions are time-bound statements; the talk is not authorization to replicate account or billing arrangements.
  • Anthropomorphic language about agents is the speaker's framing, not evidence of consciousness or emotional needs.
  • The source does not establish that all personalized factories are nonportable, all creative AI output fails, or any future prediction came true.
  • Promotional interlude cues 225–235 are omitted from the bundled transcript; the remaining cue numbers retain their source positions.
  • This is a spoken transcript; unseen slides, linked repositories, and third-party research were not independently inspected. No outcome beyond the described talk is established.

skills

talk-yegge-ai-powered-factories

tile.json