A senior Anthropic engineer just dropped 11-page PDF on "Loop Engineering" for agentic systems. The shift: you stop prompting the agent. You build the system that prompts it instead. Schedule → Discover → Build → Verify → Repeat Every loop runs one turn, five moves: • Discovery: it finds its own work - failing CI, open issues, recent commits - instead of being handed a list. • Handoff: each task gets an isolated git worktree so parallel agents don't collide. • Verification: a second agent, told to assume the code is broken, reviews the first. The "thing that can say no." • Persistence: results get written to disk, never left in a context window that gets flushed. • Scheduling: an automation wakes it on a timer. That's what makes it a loop. The key insight: an agent grading its own work always praises it. This 11-page PDF changed how I'm building agentic systems today. Read it now, then explore the article below.

@0xCodez Schedule → Discover → Build → Verify → Repeat is essentially the best work cycle

@0xCodez learned from it

@0xCodez interesting thing for reading thanks, Codez

@0xCodez very useful read ! thanks for the share Codez !

@0xCodez @grok help me find a link to this paper

@0xCodez Ya bro he got all of this information from me. Made this system months ago. Enjoy. github.com/MetaGates/team…

@0xCodez As a beginner what is stopping putting this PDF into an AI and having it build its own loops? What input do I actually need to do? It seems the user is superfluous. Am I wrong?

@0xCodez loop my way there

@0xCodez Looping is fine But how does the reasoning work to identify how to improve?

@0xCodez Why would you blow so many tokens for substandard output, good for non scalable products

@0xCodez guys, you realise this is fake right? this pdf doesn't exist and hasn't been published by anyone at anthropic. I'd bet there's something baked into the pdf to prompt claude to do some shit.

@0xCodez I agree with “the thing that can say no.” That is why how to deploy agents is probably the most important enterprise-level operation in the next phase.

@0xCodez I’ve been doing that for two years already. This dude is late to the party.

@0xCodez verification changes everything

@0xCodez Yeah. Daniel miessler designed this as PAI almost a year ago. Old news

@0xCodez this 11 page paper about Loop engineering seems like a gift for myself

@0xCodez an agent saying no is exactly what we need

@0xCodez Looping doesn’t work. It’s a total shitshow without a human in the loop for quality control

@0xCodez Wasn't wrong with time based goals either :) x.com/Its_Philip_Alm…

@0xCodez Ofc🙄 tell us something we don't know...do.

@0xCodez 有点意思 但这个loop engineering听起来像把prompt engineer卷出新高度 我连agent都还没玩明白😂

@0xCodez That’s mental illness spread by cancer company

@0xCodez It’s hidden GEM

@0xCodez Jesus Christ people are thick. I think that's my main takeaway since the AI boom.

@0xCodez someone please tell me how to do loop engineering in cursor,I feel like my tasks are too difficult for general promoted opus 4.6

@0xCodez wild that the whole unlock is just.. making it not trust itself we built confidence into these things and now the fix is building in paranoia

@0xCodez Once upon a time, people wrote scientific papers about the structure of atoms

@0xCodez That's very helpful. Thanks for sharing. I really needed this document too. ☺️

@0xCodez 11 pages when it's simply... "Do the Thing" SMH x.com/jason_haugh/st…

@0xCodez This is spot on

Our delivery comes with a genius. Go learns what you like, builds your cart in seconds, and gets it all to your door in as fast as 15 minutes.