Planners, dispatchers and schedulers
People who override the model and have never once been asked why.
A personal AI lab, built in the open
Spin up an idea, run it, show where it broke. Agents, optimization, math and sovereign AI: open, reproducible, and built to scale past the demo.
Open collaborationBring a problem
Have an operational constraint, messy real-world dataset, or production baseline to challenge? Email me at hello@gospinup.dev. Contributions and constraint overrides are credited in the spin.
Latest spin · $0.000043 per review classified
It can't hallucinate. It scored a room nobody mentioned.05
MIT Code01
Public Data02
Runs on a laptop03
01What a spin is
A spin is a finished experiment, not a blog post.
Most technical writing describes an idea. A spin is the idea actually run — with the data it ran on, the code that produced it, the number it moved, and an honest account of where it fell apart.
Every spin follows the same seven sections in the same order, so you can skip straight to the part you came for.
Section five is the one that matters. A demo that only shows the working path teaches you nothing you can use. Everything here is MIT licensed and reproducible on a laptop.
- 01
The problem, in a practitioner's words
- 02
The data, public and linked
- 03
What I built
- 04
The result, with a number
- 05
Where it broke
- 06
Run it yourself
- 07
Discuss
02The thesis
Why most AI demos never survive contact with reality.
An agent passes its eval and fails on day two. An optimization model hits its objective and gets overridden. A proof-of-concept works until the data has to stay in the country. Every spin here goes looking for that moment across AI, optimization, math and sovereign AI, with open code so you can check it yourself.
03Audience
Who this is for
I write for the people who actually run these systems:
Operations researchers
OR people who can formulate the problem mathematically but haven't shipped an agent.
Systems & agentic engineers
Engineers building agentic systems who've never had to prove a decision was feasible.
Applied practitioners
Anyone who has watched a technically correct plan get quietly ignored.
No prerequisites beyond curiosity and a tolerance for constraint notation.
04Speaking
Speaking
Where I've presented this work, with slides and video where they exist.
Available for conference keynotes, technical workshops, panels, and podcasts. If you run a meetup, conference or reading group and this would be useful to your audience, get in touch.
Sessions & slide decks05Writing
Writing
Essays and notes, published on Substack and LinkedIn, and listed here. Longer technical writeups live on this site, where the code renders properly.
essay
2026-10-05
2026-09-23
2026-09-23
2026-10-05
NetApp INSIGHT 2026: Storage Got More Exciting!
Why a storage launch is on this notebook
essay2026-09-23
The Frontier (AI) moved twice in last 3 weeks!
22 days that shaped the AI landscape
essay2026-09-23
The scraper now costs more than the intelligence!
A primer on Jev, the model that can't write a sentence, and what happened when I pointed it at live Google reviews.
Current reading & obsessionsOn the desk between spins · /now →
01hubermanlab.com Huberman Lab: Protocols Book02goodreads.com Deep Utopia: Life and Meaning in a Solved World03wired.com OpenAI Just Claimed a Huge Math Discovery. Some Academics Are Crying Foul04informs.org Branch-and-Price for Large Scale Vehicle & Crew Scheduling05github Laya, Auto-regressive System 1 decision engine
[ Newsletter ]
Get new spins by email
Roughly twice a month: one new spin, or a note on something I got wrong. No pitch, nothing for sale, no course waiting at the end of it. Unsubscribe in one click.
You're in. The next spin arrives when it's finished, not on a schedule.
Powered by Substack. Your address goes nowhere else.
↯Community
Argue with me
Every spin has a thread on GitHub Discussions. If a constraint I mined looks over-fit, if my baseline is too easy, or if you've solved this properly in production and I'm busy reinventing it badly — say so there. Corrections get credited in the spin itself.