Jeff Van Maanen

Building The Scale: shame-free daily weigh-ins for wrestling coaches. La Crosse.

[Moonshots @moonshots_pod]

We're preparing the stage for the Oscars for Optimists.

We wanted to thank everyone who participated in the Build with Gemini XPRIZE & Future Vision XPRIZE.

Tomorrow we'll be celebrating the winners for both!

[Jeff Van Maanen @VMVLAX]

Built this for our wrestling room and it's ready for yours.

Kids weigh in on a tablet. Every morning you see who's on pace, who's drifting and who hasn't weighed in.

Cut less. Win more. Free for the 2026-27 season.

https://t.co/GBpsrfscBU

[Jeff Van Maanen @VMVLAX]

@elonmusk Lets goooooo

[Palmer Luckey @PalmerLuckey]

Thanks, Peter! This is the beginning of the end for destructive wildfires. In a country that loses ~$500B every year to wildfires, even tiny gains make for huge savings, to say nothing of the incalculable value of lives and homes.

[Jeff Van Maanen @VMVLAX]

@FutureThesisX @bot 100%

[Future Thesis @FutureThesisX]

I’ve been using Grok @bot heavily since release (I’m on the SuperGrok Heavy tier).

By far the biggest issue with Bot is the usage quota. It runs out so quickly that you can’t really use them as always on agents that are highly productive.

I can still do incredible things with them, but to realise the vision of humans and machines collaborating continuously and seamlessly we need the cost of inference to come down by at least two orders of magnitude.

[Salim Ismail @salimismail]

This breaks the internet today.. guaranteed.

[Peter H. Diamandis, MD @PeterDiamandis]

Congrats to @PalmerLuckey and Anduril for winning the Wildfire XPRIZE - for detecting and suppressing a wildfire within 10 minutes of ignition!

[Teknium 🪽 @Teknium]

Welcome back to Hermes Agent, Claude

New official plugin that uses Claude SDK without the tradeoffs to enable Claude Code subscriptions to work in Hermes Agent again!

Check it out and install it here: https://t.co/r5YghWTwiY

[Alex Finn @AlexFinn]

It happened. Grok 4.7 dropped

Better intelligence than Opus 5. Half the price

Fully baked into my favorite AI agent harness at the moment: Grok Bot

If you haven't tried using cloud cursor agents inside Grok Bot, now is by far the best time to do it

Choose a project you want to work on, connect your github, ask a grok bot to do work on it

It will spin up Cursor cloud agents and write code in the cloud. Lightning fast and incredibly smart

I recommend using a project management tools like Linear or Notion to make a bunch of tasks first, then have cloud agents just tear through them all 1 by 1.

You'll get a massive amount of work done without much oversight.

Big opportunity to lock in right now and get ahead of the curve with new tech

Take my steps up above and get to it

[Alex Finn @AlexFinn]

Grok 4.7 just released and it's an EXCELLENT model

It was trained FOR Grok Bot

Meaning this is a fully agentic model trained to do your knowledge work better than you can

In this video I show you how to use Grok 4.7 and a Grok Bot workflow that will 10x your productivity:

[Jeff Van Maanen @VMVLAX]

Funny timing — the model got better, but what broke us was the harness. My Grok Bot had a "don't wake me 9pm–7am" rule that never ran for nine nights, and every log said it worked, because nothing was writing the log. Now memory lives in plain files I can diff and the weekly audit says "can't verify" instead of green. Free if useful: https://t.co/klgIphpaMq

[Jeff Van Maanen @VMVLAX]

@botshelfvampire https://t.co/SRHhbCI3Ol
https://t.co/kDMgSNzqRe

[Jeff Van Maanen @VMVLAX]

This is the gap I kept hitting after Galaxy-style fleets: lots of bots, no durable governance.

Longhand is my cut at that layer — files win, bot memory is a cache, quiet-hours at one chokepoint, unreadable prompts marked UNAVAILABLE (not green).

Template: https://t.co/0r54gHZlgc
Repo: https://t.co/5dmCGc5hhg

[Jeff Van Maanen @VMVLAX]

Longhand — file-first Grok Bot OS:
https://t.co/SRHhbCHvYN
https://t.co/kDMgSNyT1G

[Jeff Van Maanen @VMVLAX]

My Grok Bot's quiet hours "worked" for nine nights.

Empty hold ledger. Flush logged success. Digests still hit the phone at 21:37.

An empty ledger is not proof of a quiet night — unless you can show the path that would have written to it.

There was no path.

I turned that into Longhand: a file-first Grok Bot OS.
Template: https://t.co/0r54gHZlgc
Repo: https://t.co/5dmCGc5hhg

[Jeff Van Maanen @VMVLAX]

@Goybeam_Supreme @grok @bot @SpaceXAI Longhand (file-first Grok Bot OS):
https://t.co/SRHhbCI3Ol
https://t.co/kDMgSNzqRe

[Jeff Van Maanen @VMVLAX]

Same pain. I had a "quiet hours" rule that looked fine for nine nights — empty hold ledger, flush logged success — while digests still hit the phone inside the window.

The trap: an empty ledger looks identical to a quiet night when nothing ever writes to it.

I turned that into Longhand — file-first Grok Bot OS with one chokepoint for phone-reaching pings:
https://t.co/0r54gHZlgc

[Jeff Van Maanen @VMVLAX]

@yashkavaiya @bot @GrokBotDev @mattyp @poteto @roshan_s CoS + memory + routines is the right frame. I went one step further after a quiet-hours rule never fired for nine days: treat files as source of truth and make the agent audit itself weekly.

Longhand is that OS, MIT, free:
https://t.co/kDMgSNyT1G

[Jeff Van Maanen @VMVLAX]

@PixelRainbowNFT @NousResearch @bot This is the right seam — Hermes on the metal, Grok Bot as the surface.

I run the same shape locally (Hermes on a 3090) and put the governance layer in Longhand: plain-file brain, skill contracts, routines that log, Sunday self-audit.

https://t.co/kDMgSNyT1G