The notebook
Notes
Field notes from real jobs. Field-tested means it shipped and survived. On the bench means I'm still collecting evidence — published anyway, because a real workbench has unfinished work on it.
Why the John Deere Right-to-Repair Settlement Matters to Independent Diagnostics
A workshop view of the John Deere right-to-repair fight: fault codes, pairing, limp mode, dealer software, and why repair access matters far beyond tractors.
The Four-Machine Test: Emergence, Evolution and the Cage
Four cheap mini PCs are enough to test useful AI emergence, but only if the colony is caged, logged and judged by working artifacts.
The System Prompt Is Not the Safety System
Autonomous AI does not need monster stories. It needs the same thing every serious machine needs: hard limits it cannot talk around.
Fable 5 Was Cancelled, So I Want More Switches
If a frontier model can disappear overnight, and if manipulation would not announce itself, local backups and physical control are not paranoia.
The Future Is Not One Giant Model
The Fable 5 wobble is a reminder that serious AI builders should not rent their entire nervous system from one remote model provider.
The Human Brain Was Not Built for the Infinite Feed
John's field note on information overload, family, trust, local AI, and why private AI systems are not about hiding — they are about keeping human agency alive.
Body-Adjacent AI Needs a Kill Switch on Our Side
If AI is moving into glasses, cars, tools, workshops and eventually body-adjacent interfaces, the off switch cannot live in the cloud.
Fable 5 Feels Like Two Models in One
An operator's observation from tool-heavy sessions with Claude Fable 5: it behaves less like a single plain model and more like a front controller in front of a reasoning engine — and that pattern is exactly how we should be building our own agent systems.
Why I Rebuilt My AI Agent’s Memory Instead of Blaming the Model
John explains why messy long-term AI agent memory caused false positives, bad routing and unreliable behaviour — and how cleaning memory made Hermes Agent safer and more useful.
Why Prompting Is Becoming Spec Writing
Chat prompts are questions. Agent prompts are work orders: scope, authorization, boundaries, acceptance criteria. The skill that matters now is the one engineers always hated doing.
The Real Problem With AI Agents Is Context Selection
Agents don't usually fail because the model is too small. They fail because the wrong slice of reality was in the window when the decision got made.
What Auto Diagnostics Taught Me About AI Debugging
Twenty years of fault-finding on vehicles maps almost one-to-one onto debugging AI systems: observe, hypothesise, test one variable, verify the fix — and never trust the fault code.
Building Local AI Like a Workshop, Not a Toy
Eight 3090s, a Proxmox host, and a rule: every machine earns its keep. Local AI infrastructure built like workshop equipment — bought for jobs, not for benchmarks.
Why Useful Agents Need Verifiers, Not Just Bigger Models
The cheapest reliability upgrade in any agent system isn't a smarter model — it's a second pass that checks the first one's work before it leaves the building.