Hacker News: stefanve

I Built a 2300-File Research Workspace with AI: Here's What It Does

stefanve — Sun, 08 Mar 2026 16:56:07 +0000

Article URL: https://medium.com/@stefanvanegmond/i-built-a-2300-file-research-workspace-with-ai-heres-what-it-does-ab54d5a985c4

Comments URL: https://news.ycombinator.com/item?id=47298835

Points: 1

# Comments: 0

New comment by stefanve in "People Who Won't Be Replaced by AI Are the Ones Who Outpace It"

stefanve — Thu, 19 Feb 2026 21:10:34 +0000

so far so good:)

People Who Won't Be Replaced by AI Are the Ones Who Outpace It

stefanve — Thu, 19 Feb 2026 20:57:01 +0000

Article URL: https://medium.com/@stefanvanegmond/the-people-who-wont-be-replaced-by-ai-are-the-ones-who-outpace-it-116ca0425c81

Comments URL: https://news.ycombinator.com/item?id=47079233

Points: 2

# Comments: 2

Structure Beats Prose: Specs for Coding Agents That Work

stefanve — Tue, 10 Feb 2026 07:20:13 +0000

Article URL: https://medium.com/@stefanvanegmond/structure-beats-prose-specs-for-coding-agents-that-actually-work-e035929b0f3d

Comments URL: https://news.ycombinator.com/item?id=46956405

Points: 1

# Comments: 0

Coding Agents Need More Than Examples. They Need Guardrails

stefanve — Tue, 03 Feb 2026 17:28:14 +0000

Article URL: https://medium.com/@stefanvanegmond/coding-agents-need-more-than-examples-they-need-guardrails-1b3f71bc2c1d

Comments URL: https://news.ycombinator.com/item?id=46874011

Points: 2

# Comments: 0

New comment by stefanve in "I Built a 2300-File Codebase with AI. How I Prevented Architectural Drift"

stefanve — Tue, 27 Jan 2026 15:25:11 +0000

I spent 1500 hours building an app with AI coding assistants. The code compiled and tests passed. But it was full of silent issues: patterns violated, conventions ignored, technical debt accumulating invisibly. This is the hidden cost of AI-assisted development. The code works today. The codebase becomes unmaintainable tomorrow. I built a tool to prevent this, then benchmarked it across 5 different AI models. Key finding: top-tier models don't need help writing working code. They need help writing code that belongs. Full analysis and open source tool linked below.

I Built a 2300-File Codebase with AI. How I Prevented Architectural Drift

stefanve — Tue, 27 Jan 2026 15:25:11 +0000

Article URL: https://medium.com/@stefanvanegmond/i-built-a-2300-file-codebase-with-ai-heres-the-jig-i-built-to-prevent-architectural-drift-56453fe2d5b4

Comments URL: https://news.ycombinator.com/item?id=46781219

Points: 1

# Comments: 1

New comment by stefanve in "Apple Photos phones home on iOS 18 and macOS 15"

stefanve — Sun, 29 Dec 2024 09:36:28 +0000

I wonder how much of the people that are being upset about an option that recognizes landmarks and that is build in a privacy minded way are using chatGTP, windows, google or socials

The writer of the post admits that he doesn’t understand the very tech lite explanation of Apple nor read about Apple AI in general and the way it was setup. A lot of people are upset that (maybe, but unlikely) Apple knows that you made the 10 billionth photo of the Golden Gate Bridge. But continu using all sorts of services, living in sec camera heavy cities etc.

Not an apple fan but from all big tech corporations they have the least amount of interest in mining user data from a business perspective

New comment by stefanve in "Ask HN: Is there any software you only made for your own use but nobody else?"

stefanve — Fri, 05 Jul 2024 14:20:53 +0000

created a car selection script, that collected data from several platforms, reliability data, occasion adverts, safety data, reviews and specs.

I used it to buy a secondhand car based on several parameters (price, doors, features, reliability etc) and by assigning weight to the parameters

Python in total a day or so of work (including tweaking and adding features) only used it a couple of times to buy a car

New comment by stefanve in "CriticGPT: Finding GPT-4's mistakes with GPT-4"

stefanve — Fri, 28 Jun 2024 07:38:27 +0000

there are many types of a wrong answer, and the difference is based on how the answer came to be. In case of BS/Hallucination there is no reason or logic behind the answer it is basically, in the case of LLM, just random text. There was no reasoning behind the output or it wasn't based on facts.

You can argue if it matters how a wrong answer came about ofc but there is a difference

New comment by stefanve in "Cyber Scarecrow"

stefanve — Tue, 18 Jun 2024 09:49:35 +0000

I get the idea but the "science" is based on reports it doesn't look like this has been tested with actual malware. Would be interesting to know how well it works

Also make it OSS and ask for donations. Not sure what your feature earning model is but is seems easy to replicate and as point out several times right now it asked to blindly thrust you

New comment by stefanve in "Tesla's FSD – A Useless Technology Demo"

stefanve — Sat, 15 Jun 2024 08:33:17 +0000

While it sounds impressive from a technical point of view it doesn’t say anything about how much better it is or if it is now truly full self driving.

New comment by stefanve in "Patagonia founder gives away the company"

stefanve — Thu, 15 Sep 2022 12:42:28 +0000

I think the 17m is only for the 2% of the voting shares the rest is political donation and thus taxfree

New comment by stefanve in "Ladybird: A new cross-platform browser project"

stefanve — Mon, 12 Sep 2022 13:29:38 +0000

you properly mean the average user. The average user would not download this browser. so the user of ladybird might want to live with the shortcomings. That being said , Firefox still exists and is a very good cross platform OSS browser

New comment by stefanve in "BMW F Series Gear Selector, Part Two: Breakthrough"

stefanve — Thu, 30 Jun 2022 07:37:19 +0000

e91 was a very nice car to drive the f36 is the first car that I modified because it didn't driver that well. changed the suspensions and now drivers in some ways nearly as good as the e91 and in other ways better

New comment by stefanve in "BMW F Series Gear Selector, Part Two: Breakthrough"

stefanve — Thu, 30 Jun 2022 07:34:16 +0000

I owned a 325xi e91 and didn't experience any turn signal problems, same for my current 420i F36. Lot of German cars (VW, Audi etc) have a two stage turn signal. A light bump will do 4 or so signals and automatically stops if you move the stalk all the way it wil stay there until you move it back.

to me the sound is loud enough to hear it but not so loud it is annoying

New comment by stefanve in "Read Easy – Dyslexia, ADHD, and SSS Helper"

stefanve — Thu, 16 Jun 2022 10:22:11 +0000

I have tried a bunch of these fonts and other tools. But for me zooming in, making the font bigger and/or using reading mode is the best solution for me. By making the font bigger the sentences get shorter and it easier to read.

New comment by stefanve in "Write documentation first, then build"

stefanve — Wed, 15 Jun 2022 13:14:36 +0000

I can't remember the last time I worked somewhere where nobody wrote down anything. Sure sometimes not enough and sometimes too much. But in general people write down specs in, stories, tasks, pitches, presentations, wireframe, slick UIs etc. Even for my own play projects I generally start out with an outline of some sorts.

but total spec upfront brings back bad memories from the waterfall world. But I have also had discussions with folks who thought that agile means not specs/documentation at all

New comment by stefanve in "DuckDuckGo faces controversy over tracking agreement with Microsoft"

stefanve — Mon, 30 May 2022 12:59:02 +0000

it's not perse about truth but it is about facts, I personally don't have a problem with downgrading lies or denies of facts. In the end of the day a algo is ranking pages, algo's can be manipulated or gamed and they are not only by lot's of marketeers but also by government agents.

While I don't believe a company should decide if an opinion about gun controle should be rankt or not. but if someone is claiming that these are just crisis actors and nothing happend I don't see a problem to block it. After all we are used to blocking spam these things are spam. I don't see a reason to treat this as any different just because they are selling political lies instate of product lies (and a lot of times these people are selling products in the end)

these kind of lies are eroding democracy and should be treaded the same way as some one yelling fire in a cinema

New comment by stefanve in "I'm tired of Google's business products"

stefanve — Thu, 14 Apr 2022 13:48:12 +0000

I worked a lot with both MS online offerings and google's. My own money go's to MS when it comes to documents etc . It is better than the offering of google especially if you need a bit more the the basics. Support is also better.

I don't thing that google office suit is bad just that MS is better

That being said I'm contamplating to stop with MS as I have less needs for a good office suit and just use what apple offers