runesleo

OpenAI added 62,500 credits to my Pro account today. They can be used on both Work / Codex and expire at the end of the year.
After thinking about it, there’s absolutely no need to conserve them.
Once the weekly Work / Codex quota is used up → keep burning credits.
Once the credits are used up → keep using GPT-5.6 in Chat.
The goal is simple:
Turn all 62,500 credits into real output.
Perishable computing power is wasted if you don’t use it. 😁

I’m becoming increasingly convinced that the best way to use AI in the multi-model era is not for one person to manage five AIs simultaneously, but to keep just one front door.
Today I fully streamlined my AI workflow around one principle:
Single Front Door for AI
ChatGPT is my brain and hands, while all other AIs are its tools.
I now use Grok, Grok Bot, Grok Build, Claude, Cursor, and Codex simultaneously, including Browser / Computer Use, MCP, my local Mac, VPS, phone, and more…
But I don’t want to switch back and forth between all these things every day.
Because once you have too many tools

Last night I minted 5 Merkel Koalas.
All of them used the fastest unlock option of 420 blocks:
#1831 / #239 / #1680 / #7967 / #7178
All 5/5 have now been REVEALED.
Let me make my stance clear first:
I will not mint a 6th one.
This is also not a call for everyone to mint.
I minted only the first one at the beginning because I wanted to use real money to fully run through the CRC-21 mechanism from commit → reveal → vault → unlock.
After the first one was successfully verified on-chain, I continued my DD.
Later, the core developer also publicly stated that LEAF (CRC-20) and Merkel Koalas (CRC-21)

After seeing too many new Bitcoin L1 projects lately, I increasingly feel that what’s easiest to overlook right now isn’t “how to buy,” but:
how to sell after buying?
The supply of new protocols, new assets, and new launchpads is growing too fast, but the liquidity infrastructure has clearly failed to keep up.
At mint, every project looks incredibly attractive:
first meme, first DeFi, Bitcoin native, low market cap, early…
But once you actually buy in, you discover:
there’s no AMM, no depth, no aggregated liquidity, and sometimes even placing an order requires carrying out on-chain operations

ChatGPT is increasingly starting to feel like a “personal workbench.”
When I opened the app today, the homepage immediately surfaced several things that might need attention based on my connected Gmail, Calendar, and GitHub:
The remote Mac may be experiencing instability
A calendar event that has already gone awry
A PR that hasn’t been reviewed yet
What’s interesting isn’t that “ChatGPT can read these apps”—it has been able to do that for a long time.
The real change is that it’s starting to proactively determine “what might need your attention right now,” instead of waiting for you to ask fir

Today I finally got something I had been thinking about for a long time working end to end: using Grok Bot + two Macs to build a multi-device network self-recovery system.
My daily environment is no longer “one computer,” but two Macs, Tailscale, VPN, Grok Bot, Claude, Codex, Cursor, and a bunch of long-running Agents and automations. This also creates a real problem: if a machine’s network, proxy, Tailscale, or Agent fails while I’m away from the computer, the AI may know how to fix it, but it cannot get in either.
So the core problem I wanted to solve today was not “how to make computers nev

A friend saw two find-the-object mini-games at a shopping mall and casually used them to test AI's visual capabilities.
Find the marshmallow among the seals
Find the lamb among the Santas
ChatGPT: Took 9 minutes to think, and found both.
Doubao: Couldn't find the marshmallow, then very confidently...
Photoshopped two marshmallows directly into the original image, circled them, and told me that was the answer 😂
AI's most hilarious moments aren't when it gets the answer wrong.
It's when it changes the question just to prove it got the answer right.

Today, ChatGPT’s “task continuity” problem hit me hard.
It wasn’t that the quality of a particular response declined, but something more troublesome:
Some conversations that had been running for a long time suddenly stopped responding;
Tools / MCP that had been working normally suddenly failed to connect in some sessions;
Although switching to a new conversation let me chat again, the task context, execution state, and progress were all very likely to be lost together.
I spent a large part of today not “getting things done,” but restoring “what AI had originally been helping me do.”
This sudde

Lately, I’ve been feeling this more and more strongly:
As the natural-language interfaces of AI Agents like Codex and Grok gradually mature, and connect to your own local workstation, browser, various tools, and the most powerful models in the cloud—
ordinary people are actually getting their own “AI operating system” for the first time.
You no longer need to open a dozen different apps, remember a bunch of commands, or manually string together workflows.
You only need to say:
“Take care of this.”
Then the Agent calls on your computer, code, browser, APIs, and models to complete the task step

Recently, I keep running into this when ChatGPT runs long tasks.
The most frustrating part is that the frontend says “streaming has been interrupted,” while the backend task is sometimes still running.
The more heavily I use agents, the more I realize that how smart the model is is only part of the equation.
Connections, reconnections, state recovery, and observability for long-running tasks are what truly affect the day-to-day user experience.

I’ve actually been using ChatGPT’s Remote Control for quite a while.
I used to think of it simply as a very convenient feature:
When I’m out, I can open my phone and continue taking control of Codex running on my Mac at home.
Send tasks from my phone → execute them locally on my Mac → check progress at any time → continue adding instructions.
But recently, as I’ve brought more and more work into AI, I’ve truly realized what this means.
Once AI can connect to computers, files, terminals, browsers, and various workflows, your phone is no longer really a “device for running AI.”
It’s more like a

I only discovered today that ChatGPT Pro actually comes with 100GB of file storage.
I’ve been using it for so long, yet I’ve only used 315MB.
This made me realize that I’ve always treated ChatGPT as “compute” and a “chat box,” but never truly as a long-term workspace.
So I reorganized my AI storage architecture:
Mac: execution
Local large drive: raw data / cold archives
Google Drive: official documents editable by humans
ChatGPT Library: a knowledge layer that AI can read, search, and reuse over the long term
The point isn’t to fill up the 100GB.
It’s to put the things that are truly worth hav