mcboom retweeted
Remember how everyone said Project Jupiter was cooked and that the $ORCL Force Majeure somehow popped the AI bubble…?
Well our friends over at the Santa Fe New Mexican just reported that the new hearing officer appointed in the Air Quality permitting proceeding is retired federal judge James O. Browning
A Republican nominated by President George W Bush in 2003 to the US District Court in NM
I’m pretty sure this guy isn’t phased about some woke protestors trying to prevent a $165Bn AI Factory 😂
His appointment also comes immediately after the New Mexico Supreme Court lifted all emergency stays, signaling a clear judicial mandate to advance the regulatory process decisively
The air quality permit is going to get approved, the gas line will be get built (or re-routed around State land), and Project Jupiter is going to go-live as planned and on-time. $ORCL
santafenewmexican.com/news/l…
mcboom retweeted
The hyperscalers have bet big on reactors. Next they'll have to lock up the fuel, and the first one to move gets the last good price.
mcboom retweeted
Yes, a bank run is coming.
Get your cash out of Bank of America while you still can.
mcboom retweeted
Your CLAUDE. md is making Opus 5.5 DUMBER.
Lines like "double check your work" and "think step by step" were written for OLD models.
On Opus 5.5 they slow it down and make the output worse.
Lance Martin at Anthropic just shipped the fix.
Open Claude Code and type:
/claude-api prompt-audit
It scans your skills, CLAUDE. md and prompts.
Cuts the lines that hold the model back.
Rewrites the rest with the official Opus 5.5 guidance.
5 minutes. Run it TONIGHT.
Bookmark this before you forget.
mcboom retweeted
BREAKING 🔥: An upcoming $500 ChatGPT Pro Max plan will likely be powered by @cerebras infrastructure.
> So far, "Fastest Work and Codex" are the only changes in the plan description compared to ChatGPT Pro.
> Usage limits could still be higher as well 👀
* The plan price is $500; it shows $600 due to VAT.
OPENAI 🔥: The upcoming ChatGPT Pro Max plan will cost $500.
So far, this will be one of the most expensive AI subscriptions available on the market.
I hope "it will be worth the wait" 👀
mcboom retweeted
useful tip for Opus 5.5:
run “/claude-api prompt-audit” in Claude Code. this checks you skills, agent.md, Claude.md, prompts and removes anti-patterns that hobble frontier models.
i updated the skill w/ the latest Opus 5.5 guidance.
mcboom retweeted
FUCK.
Been a while since I posted something this interesting.
Today, QQQ was
▪ Up at least +2.75%
▪ On a day VXN (VIX for QQQ) was up at least +1.00 point
And how has QQQ done 2 weeks and 1 month after every similar event in the past?
Hahahahahahafuck.
mcboom retweeted
If you are confused about why 𝗝𝗲𝘃 is being called the "Internet" moment for the AI industry. This is 100% worth your time. In fact, you should watch it:
It tells LLMs what to do next, in milliseconds & at almost zero cost. If you set it up correctly, you will have the AI engineer’s setup for 2028.
How to set up & use Jev (to actually get the 100x):
1. Join the waitlist; it's fairly quick: typesafe .ai.
2. Then go to Claude Code or Codex.
3. Choose Opus 5-Low or Sol-Low.
4. Copy and paste this prompt:
"[claude or codex] plugin marketplace add typesafe-ai/skills
[claude or codex] plugin install typesafe@typesafe-ai"
5. When you type /typesafe, the skill shows up.
6. Paste your API key once and click "Allow"
7. Start with $5 in free credit. It's hard to spend more.
-----
Now, here are the 3 ways to actually use Jev:
1. Jev for Linkedin
I have 38,000 connections & invitations on LinkedIn.
I have a new company to launch. I need to find a couple of hundred people to message. to do while saving time:
> Export your LinkedIn connections and invitations.
> Connect Claude to GitHub, Vercel & Apify.
> Create an Apify API key to enrich your data.
> Go to LinkedIn Settings → Data privacy.
> Get a copy. LinkedIn will email you a ZIP file.
> Open the file & find the Connections CSVs.
> Upload the files to Jev.
> Use Jev to classify your contacts.
> Review the shortlist.
2. Jev for Gmail
To go through all of my Gmail contacts and email the right people.
> Go to Google Contacts. Open Other contacts.
> Select all contacts. Export them.
> Upload the file to Claude Code or Codex.
> Use Jev to sort them into:
Keep, Review, Remove and Review everything before removing anything.
3. You got lost in Claude Code, GitHub, Vercel, Apify, Jev, Typesafe.
I feel you.
It is overwhelming.
That’s why I included the entire copy-and-paste prompt for each use case in the newsletter: ruben.substack.com/p/jev
A 45-second TL;DR by @MatijaSosic.
mcboom retweeted
Okay Instinct is wild. Just found and filed for $2,182 in lost dollars owed to me. It even found my ssn proof and signed the forms using a photo of my passport for signature. 🤯 Prompt: “Search every state's unclaimed property database for money owed to me. Use my full name and every city I've lived in. For anything you find: tell me what it is, who owes it, and how much - then file the claims end-to-end. Walk me through any documents or signatures you need from me, and don't stop until every claim is submitted.”
Thanks to @peterpham for the tip.
mcboom retweeted
OpenAI quietly showed how to run 1,000+ agent loops from one prompt on GPT-6 Astra
3-page field guide on exactly how to turn Astra from a chatbot into a fleet, along with the COMPLETE OPERATOR BLOCK you paste once
here is how you set it up:
1. the 3 lanes of action Astra actually sorts every task into (reversible → just do it, consequential → do the work then ask once, irreversible → stop) and why the approval line is reversibility, not task size
2. the operator block that fixes the 5 default behaviours OpenAI documented themselves: stalling on "can you…", over-testing trivial edits, answering in Markdown, obeying stray lines in skill files, and delegating less than you want
3. the effort ladder (low / medium / high / xhigh / max, no "none") with configuration_update to raise it for one hard turn without breaking the cache prefix
4. the 3 work orders that use the whole model: research memo with traceable citations, Codex refactor in a worktree with a green-suite gate, computer-use task with a hard boundary on anything a customer would see
5. the edge test that finds the parallelism hiding in your own workflow, plus the 3 things that break every fleet: self-agreement (verifier on clean context), file collisions (one worktree per worker), the 272K cliff that reprices your whole request
6. the one Responses request that runs 20 workers behind an async tool with a verifier that re-runs the test instead of asking the agent if it's done
this is the exact setup:
mcboom retweeted
How to use these Prompts
- Open ChatGPT (work, not the free version)
- Select GPT-6 Astra
- Copy one of the prompts
- Replace anything inside [brackets] with your own details
- Give Astra access to the apps, websites, or files it needs
- Review important actions before it sends, buys, deletes, or changes anything
- Let it complete the task, then ask it to save the result as a file, report, spreadsheet, or checklist
mcboom retweeted
Bro had $100mm in (presumably) levered futures betting on a weak NFP, consensus was 260k and the print was 140k…he’s gonna be wayyyyy more excited than that
