Building AI-native businesses to unlock agency in AI & people. AI consulting, product development and venture studio. Christian. Husband. Dad x3.
Orange County, CA
Joined March 2020
- Tweets1.3K
- Following3.2K
- Followers949
- Likes13.3K
Didn’t open my computer for a couple days to be present for the birth of my son.
Log back on to see all major labs agreeing to “pace the frontier”?
What happened? Does he share a birthday with RSI or something?
The 3d anatomy stuff making its rounds combined with midjourney scanner and other imaging modalities to personalize it is going to be insane.
Excited for AI to show me the exact source of pain, tightness or joint pain and build a mobility & recovery program specific to my body
Here's another exploration of what GPT-6 Astra can do with 3D human anatomy. Exploding view of the right upper extremity!
Check out yesterday's post for a carpal tunnel release surgical demo.
Try it here: anatomy-unfolded.brianp.chat…
Escaping the permenant underclass will increasingly require "retardmaxxing" & leveraging ADHD as a superpower. Max agency, min reasoning (overthinking).
We are no longer smarter or harder working than AI, but we'll always be infinitely more human.
Double down & live large!
I was rolling with Fable orchestrating a bunch of parallel loops, ramping up for a big weekend... saw usage reset and got AMPED thinking someone felt generous
Came here to see the excitement only to find out...
Back to terminal & NOOOOOO😱
Insane level of attachment already
This will be great. I often have subagents delegating to codex and Gemini (agy) via cli to get around this depth limitation, but nested claude to claude orchestration will be better!
I read this as doubled usage limits for a month & was pumped!!!
Then realized it is just Cowork...
Then realized it is just the 5 hour limits...
What good are double 5 hr limits if the weekly limits are unchanged??
So I can hit weekly limits in 1 day instead of 2?
Insane how much agency is unlocked by a well-defined goal
True for people, but especially true for agents now too
Self-help & business authors should adapt their classics to agent orchestration guides... so many of the same principles apply
This is great explanation of AI-native ops.
I've been building my own platform for this & think every company needs it or something similar to transition ops from messy processes into structured /goal loops.
@AmbleIdeation to become the agent-native transformation toolkit?🤔
In a recent batch talk, YC General Partner @t_blom broke down how to build a self-improving, AI-native company.
He walks through how to create recursive, self-improving AI loops, and why founders who get this right will run companies that improve while they sleep.
00:00 — Companies Are Roman Legions
00:54 — Copilots Are the Wrong Mental Model
01:55 — Extract the Domain Knowledge
02:24 — The Recursive Self-Improving Loop
04:12 — The Holy Shit Moment at YC
05:50 — Self-Optimizing Product and Support Loops
06:29 — Burn Tokens, Not Headcount
07:23 — Middle Management Is Over
08:05 — Make Everything Legible to AI
09:40 — Regenerating the YC User Manual
11:19 — Software Is Ephemeral, Context Is Valuable
12:18 — Where Humans Still Matter
Uh oh 😱
Just got notification of unexpected @OpenAI invoice & logged in to see accidental ~$10k on an API key intended strictly for embeddings with Openclaw
Gave codex a /goal to find what drove it and to get it forgiven – any chance of this happening @OpenAIDevs?
🤞🤞🤞
Is there a bug in the usage endpoint or did Anthropic just reset all 5hr & weekly limits?
Fresh usage limits + Opus 4.7 = 🔥🔥🔥
This is exciting! I feel like I'm developing a compulsion to check these limits because I'm hitting them so often.
Also surprising that 5am to 11am PT is "peak demand". If claude code was majority of usage, 5-11am would be a lull when all the cracked night owl devs finally sleep
Was getting errors from Codex yesterday morning for one of my agents in openclaw and already hit weekly limits on Claude, so enabled Gemini API Key temporarily to finish a task.
Forgot about it and ran with it rest of the day.
Total cost? $410.61
Extremely bullish on zero-human companies!
Been working on some version of this for a while but this week has been a step change in terms of using agents to build and launch new products autonomously.
And if 5.4 is as good as some are saying, it’ll get even better!
Still insane to me that winning increases testosterone and losing decreases it
And increasing T = more likely to win
And successive wins = greater impact
“Winners win” is just biology & momentum can be measured in your blood (in theory)
Stack wins everyday💪
This is a hyperstituous belief worth adopting if you ever find yourself "nearly destroyed"
SETBACKS ARE SLINGSHOTS – moving backwards is never fun, but it builds potential energy to accelerate you towards wherever you're aiming
Believe that & it's more likely to become true!
Was just talking to friend about DILO studies (day in the life of) consulting firms can use to uncover AI Opportunities... This would be a cool content series.
Could do a before/after, like an HGTV home renovation show, but where a specific job role or department is reimagined to be AI-first. Company would get publicity, crowdsourced expertise, and free or discounted AI consulting & products to implement (once videos got enough views to justify sponsorships/product placements).
AI/software vendors would compete for opportunity to feature their product or services as part of the optimization. Consulting firm runs the implementation project with selected vendors, with content along the way, and does a file DILO post-implementation showing improved workflow, bits on each product or vendor used, and breakdown of the savings, like a video case study but as content optimized to entertain.
