hugomelis.nl/experiments
3 AI Experiments
That taught me more about myself than about the tool.
Hugo Melis · 2026
hugomelis.nl/experiments
3 AI Experiments
That taught me more about myself than about the tool.
Hugo Melis · 2026
Day 7
Feb 24, 2026
Running a Local Model to Allow
My Thoughts to Flow Freely
I struggle with the amount of data trail I leave online. With the rise of AI my privacy paranoia has gone up. I am deliberate about what I feed to online AI tools, and what things I only tell my local models.
Day 7
What was attempted
THE PROBLEM
Hugo records voice memos on walks — raw, unfiltered thinking. But sending those to cloud AI means his most private thoughts become someone else's data.
THE APPROACH
Local Ollama models + OpenWhisper transcription. Nothing leaves the machine. Privacy becomes infrastructure, not policy.
Day 7
Key Insight
Key Insight:
Privacy is not about neighbors looking through the window. It is about how freely I allow my thoughts to wander.
Maximum experimentation, minimal exposure. Upgrading infrastructure to run bigger models allows the freedom to think without surveillance.
Day 10
Feb 27, 2026
Troostwijk Auction Search Sucks,
Let's Fix That
We are building a house and need high ticket items. The Troostwijk app's limited search creates pricing inefficiencies. Misspelled listings like "mutliplank" instead of "multiplank" meant deals went unnoticed.
Day 10
What happened
TRIGGER
Misspelled auction listings meant deals went unnoticed by other bidders.
BUILD
Used Codex /plan mode + Docker. A bot that finds partial matches and misspellings.
RESULT
Found auction deals invisible to normal search. Working prototype in ~2 hours.
Day 10
Key Insight
Key Insight:
As AI gets better, my inspection threshold lowers. Leverage increases. Visibility decreases.
BEFORE
Review every line of code.
Question each architecture decision.
Maintain full comprehension.
AFTER
Skim the plan. Trust the structure.
Contain the environment, not the decisions.
Surrender comprehension for speed.
Day 19
Mar 8, 2026
Building My Own CRM:
Can the Vision Survive?
I always wanted my own personal CRM. Not an enterprise CRM, but something shiny and fun to use. After years running an app development agency, the vision never survived the translation layer.
Day 19
What changed
THE OLD WAY
Rough idea → wireframes → designer back-and-forth → developer builds → realize it's not right → more revisions. Weeks pass. Budget runs out.
THE AI WAY
Skip wireframes entirely. Claude Cowork in Figma produces full-fidelity designs in a few iterations. The vision holder becomes the builder.
Day 19
Key Insight
Key Insight:
The cost of translating an idea into a working product has dropped to almost zero. We don't know what this means yet.
The question shifts from "can I build it?" to "is this what I actually want?"
The Hidden Pattern
Every experiment follows the same loop:
01
Control
Set up the experiment. Define the question.
02
Confrontation
Run into what AI actually does vs. expectations.
03
Self-Knowledge
Learn something about how I work, not just the tool.
hugomelis.nl/experiments