HYPERREAL ENTERPRISES LTD

Is your AI solution working for you?⊕ Raven

A Minard-style stream of one month of an operator's work with AI agents, 22 August to 21 September 2026, coloured by the kind of work (perceive, believe, evaluate, select, act, assurance, coordination), with annotations: 30 Aug facade found; 05 Sep flying, theory work set to run; 10 Sep topology campaign, automation that went well; 13 Sep codex-26 builds unrequested bookkeeping.

⊕Me running agents like a boss.
Voxterm on a phone: push to talk, speech transcribed on the server, and a column of live AI agent seats (Claude, Codex, Kimi) each marked invoking, above a tree of the processes they are running. I thought I was getting really good at agentic coding. I built my own system for automating agents. Then I realised the agents weren't building what I asked for. Does this sound familiar?

More automation brings more chances for agents to go off and build things nobody requested or wanted. When a request is well specified, things go better; when it isn't, the agents make up the details as they go, and you end up redoing the work. Automation won't, by itself, solve the problems you want solved.

Have a look at your logs. This Python file reads your Claude Code and Codex logs on your machine. No network, no language model, standard library only. It is short enough to read before you run it. It will tell you what kinds of request you make, and how many credentials are sitting in the logs (it doesn't print them).

curl -O https://hyperreal.enterprises/downloads/xiaoxiang-local.py
shasum -a 256 xiaoxiang-local.py
less xiaoxiang-local.py
python3 xiaoxiang-local.py --days 7

Download xiaoxiang-local.py · SHA-256 dae32d7e3c5b4704e732278b3e01abb07ae02078144f9e359e8be6a2b173d434

How 小象 (little elephant) reads intent

Mirrored bars on a date axis, 22 August to 21 September 2026: the tokens agents logged during each recorded gap in the operator's activity.

Each bar is a gap in my recorded activity: its width is how long the gap lasted, and its height the tokens agents logged during it.

We’re building an agentic harness for knowledge work.

⊕ A dog walking upright on two legs leads another dog, on four legs, by a leash. Decades of research on learning and on mapping knowledge domains, consulting with clients on strategic plans of action, and years of work with online communities have helped us build methods guiding and governing AI-assisted work. In 2019 we put forward a long-term vision of computational intelligence based on collective intelligence. It turns out, LLMs provide that. Now we're building a methodology for using them, based on the way collective intelligence works.

We can help you realise your vision.

There is often a gap between what you think you want and what your work actually does. Agents can widen it, or narrow it. The first step is to find that gap. Then we can help you steer.

Click through for more information.

HYPERREAL ENTERPRISES LTD is a private company limited by shares, incorporated 25 June 2019 as Company Number 634284 on the Register of Companies for Scotland.