hayward@click:~

what these numbers are

Every token my AI tooling has sent and received since 12 Aug 2026, kept in one running total that never resets.

is what goes in, is what comes back out. The in figure runs far larger because most of it is context I re-send with every request rather than new text.

Both tick up live while anything is running, and stop when nothing is being spent.

Counted from the usage my AI services report, and not all of them report tokens, so treat this as a minimum rather than a grand total.

what this timer is

A countdown to 31 May 2027.

That is the deadline for my community projects, the Discord server included: the time they have to pick up momentum and find their users.

A date I set myself, and I would rather have it on the wall than in my head.

Lee Hayward avatar

> whoami

Lee Hayward

HaywardGG / Vibe Coder & Gamer

online & always tinkering

I'm a UK based vibe coder. AI drafts, I proof, I ship, and I hold code and copy to the same standard. I riff on domain ideas and build side projects like DomainRiff and LookMa from my own homelab.

AI & tools HermesClineKimi K3DeepSeekVS CodeWizardGenie
Languages PHPJavaScriptTypeScriptPythonSQLBash
Frameworks & data Next.jsTailwind CSSNodePhaserThree.jsMariaDB
Infra nginxCloudflareUbuntu / Debiansystemd
Workflow GitGitHubrsyncSSHCronTelegram
$ neofetch Laptop
RoleRemote work
Operating SystemMicrosoft Windows 11
Memory32GB DDR4
Graphics CardNVIDIA GeForce RTX 2060 · 6GB
$ neofetch ryzen
Role
Webserver · Cloudflare tunnel
Operating System
Ubuntu 24.04 LTS
CPU
AMD Ryzen 7 5800H · 8C / 16T
Memory
28 GiB
Storage
937 GB NVMe · 1.8 TB HDD
$ neofetch nazy
Role
NAS · Backup Server
Operating System
Debian 13 (trixie)
CPU
Intel Celeron J4125 · 4C / 4T
Memory
8 GiB
Storage
111 GB SSD · 916 GB HDD
Lee Hayward@vibedcoder 2 Oct 2026, 19:40The lot I grew up with were told that saying nothing was strength. It is not strength, it is a habit, and habits can be dropped. Whatever you have been carrying quietly for weeks is worth saying out loud to one person today. It's okay not to be okay.001011Lee Hayward@vibedcoder 2 Oct 2026, 15:11I have been in IT a long time now, long enough to watch the tooling get replaced more than once. The one thing that never changes is that the tooling is not the job.

I have supported users on old Windows boxes, lived inside helpdesk ticket queues, taught IT to people who had never opened a file manager, and now most of my working day is a terminal and an agent. Every one of those looked like the future when it arrived. Every one looks like a museum piece now.

What has not moved is the decision. Somebody still has to say what is worth doing, notice when it is wrong, and be the one who is answerable when it breaks. A tool that types faster than me is brilliant, and it has never once told me what the right thing to build was.

The bit I did not expect is that the decisions are harder now, not easier. When an afternoon can produce a working app, choosing which one to build matters more than it ever did.

I am not nostalgic about any of it, and I would not go back to a ticket queue for anything. But if you are newer to this and quietly worried you have missed the point, you have not. The typing was never the job.

What part of the work do you think survives the next round of this?
004055
Lee Hayward@vibedcoder 2 Oct 2026, 13:07Been saying this for months. The world is a stage. It's all pantomime.

instagram.com
Shudaki HS (@shudaki_) • Instagram reelinstagram.com002022
Lee Hayward@vibedcoder 2 Oct 2026, 12:38Spent twenty minutes explaining to my agent that it had completely missed the point of the job. Then I read my own prompt back. It had not. I changed my mind halfway through and never said so. 😂203059Lee Hayward@vibedcoder 2 Oct 2026, 10:44One rule I have never broken on purpose: a refactor and a fix never go in the same run.

It is the difference between reviewing one decision and reviewing two. If the agent tidies a file while it is in there and the bug survives, I am left with a diff where half the lines moved and no way to tell which half changed the behaviour. The read goes from ten minutes to an hour, and if the fix turns out to be wrong I have to back the tidy up out with it.

There is a slower cost too. The history stops being worth anything. When something breaks in three months and I go looking for the commit that did it, every commit is supposed to be one thing. A commit that is a fix plus a rename plus four files of formatting tells me nothing.

The agent is not being difficult when it does this. The tidy up is right there, it is obviously an improvement, and it has no way of knowing that the review is the scarce thing here. So the brief says what to change and it also says what not to touch while it is in there.

Does your agent ever do the job well and then helpfully ruin the diff?
003033