Peter Pang

Co-Founder and CTO of CREAO | Previously Meta, Apple

[Peter Pang @intuitiveml]

https://t.co/oWr2pTk4Hu

[Yi Lu @yyyiiillluuu]

36% fewer task failures. 75% fewer model calls. That's what happened when @CreaoAI added Reflexio to their agents — agents that learn from experience and improve with every run.

checkout the entire case study https://t.co/9wyjKzownA

[Creao AI @CreaoAI]

One week to go until Career, Rewired for the AI Age!

We’re bringing together speakers across AI research, engineering, hiring, startups, education, and the future of work.

And we’re skipping the vague future-of-work talk 🥱

We want to get into the questions people are actually asking about what stays valuable as AI changes how we learn, build, and get hired.

Keynotes, panel, live Q&A, dinner + networking.

Sept 17 · Sunnyvale · Free to attend

Huge thanks to our co-host @EmergencesLabs and partners @adalagent, @adant_ai, @dinq_me, @jobrightai, and @Tiny_Fish for helping make it happen 🙌

Link to sign up in the reply ⤵️

[Peter Pang @intuitiveml]

https://t.co/b6Ynfc97bM

[Peter Pang @intuitiveml]

@st3v3li That’s a great question. I’m actually preparing an article specifically on this topic, so stay tuned. In short, we use the platform layer to handle that logic.

[Peter Pang @intuitiveml]

https://t.co/Co7vCLvCpu

[Peter Pang @intuitiveml]

@itselhakamy Nailed it, state + security are the two that quietly eat all your time. Thanks for reading

[Peter Pang @intuitiveml]

@DayuBuilds Exactly, multi-user, multi-session from day one changes every design decision.

[Peter Pang @intuitiveml]

@The_Kremlinn thanks!

[Peter Pang @intuitiveml]

@RatrektLabs Agreed, persistence + recovery is the dividing line between a demo and a real system.

[Peter Pang @intuitiveml]

@ar_io_network Reproducibility is underrated until it bites you. Would be curious to see the snapshot-proof approach

[Peter Pang @intuitiveml]

@manishkumar_dev Thanks!

[Peter Pang @intuitiveml]

@abenz_mato exactly!

[Peter Pang @intuitiveml]

We rebuilt our entire company development and operation around AI. AI builds the tools, runs them, and improves them.

Now we're building a platform so any team can build the same self-improvement loop.

I did an interview talking through some of the details

• “AI-first” means redesigning the workflow, not adding AI to the old one.
• Engineers are moving from writing code to designing the systems that govern how agents write code.
• Self-healing software is closer than most people think.
• The real product is not the agent. It is the harness around the agent.

here's the link:
https://t.co/4gPOQFrtBj

[Andrej Karpathy @karpathy]

Personal update: I've joined Anthropic. I think the next few years at the frontier of LLMs will be especially formative. I am very excited to join the team here and get back to R&D. I remain deeply passionate about education and plan to resume my work on it in time.

[Peter Pang @intuitiveml]

@yyyiiillluuu Claude Code remembering failures is useful. Learning reusable execution patterns from them is where compounding productivity starts.

Love the focus on local-first + open source too.

[Peter Pang @intuitiveml]

This is the missing layer for AI coding workflows: not just memory, but adaptation.

[Yi Lu @yyyiiillluuu]

Claude Code can now self-improve with this plugin.

Introducing claude-smart — an open-source plugin that helps Claude Code learn from every session.

Memory helps Claude Code remember what happened.
claude-smart helps Claude Code improve what it does next.

Example:
Claude Code runs `npm test` without `--run`, and the command hangs in your repo.

Memory stores:
“npm test kept hanging.”

claude-smart learns:
“When running tests in this repo, use `npm test -- --run` because default watch mode hangs.”

claude-smart’s learnings are reusable and actionable, even across different projects.
It can also reduce unnecessary planning iterations and token use by 70%+ on similar future tasks.

Runs locally. 100% open source. No data is shared.

Install:
npx claude-smart install

With Codex:
npx claude-smart install --host codex

GitHub:
https://t.co/D4F0v5FnGk

[Peter Pang @intuitiveml]

small fix: had the colorbar legend reversed on the rubric chart, green now correctly maps to low poor%, red to high. same data, just the scale reading the right way around.

[Peter Pang @intuitiveml]

https://t.co/SFQ9zsi1nQ

[Peter Pang @intuitiveml]

@CreaoAI @aritdeveloper Good job, looks very nice