allanmacgregor

Joined 13 years ago

Karma
330
Submissions
241
Member since
Jul 20
20 years turning technical bets into shipped products. CTO @ LiORA, building Bloccs. Ex-CTPO @ Humi ($140M+ CAD exit). I write http://thepragmaticcto.com
View on Hacker News

Recent submissions

Comment · 1 month ago
I thought this wasn't really an LLM sure text as input but the folks at Typesafe have been very adamant this is a new class of model not just another LLM.

Although to be fair we don't know enough about the architecture.

Comment · 1 month ago
> and ultimately they have to go somewhere; wherever the government puts them is going to object.

Back to their own country, look I am an immigrant myself but there is this massive trend of goverments developing suicidal empathy even at the cost of their own citizens; it really needs to stop.

Comment · 1 month ago
One of the things that I found extremely helpful is combining agents running different models to review and challenge the other one. Trying to coordinate both and have them have a single point of communication has always been challenging.

Radio seems like an interesting solution to the problem I'm very keen to try, this as part of my workflow.

Comment · 1 month ago
Bit of a strawman here, particularly the last one. Find a single US administration in recent history that hasn't killed thousands in the middle east? I'll wait.
Comment · 1 month ago
Sorry, but people are calling anyone who disagrees with them nazis or fascists so frequently that it is hard to take seriously anymore?

You don't agree with my exact political views?! Then you are a fascist.

Also blatant fascism? By which definition? the OP article pretty much cites a single blog post as his evidence of DHH being a fascist; but never really defines fascism.

I get it DHH is not a very likeable guy, and a bit of a prick but calling him a fascist or taking childish jabs like calling Omarchy "vibecoded scripts on top of Arch" really removes a lot of credibility

Comment · 2 months ago
Scholars in what fields?
Comment · 2 months ago
This is exactly what it is, just slop. I do agree with the intent we need someway to verify and validate agentic coding output that doesn't rely on codereviews; but this project is just sloppy.
Comment · 2 months ago
Is no one going to call out the lazy and clear AI Slop on the landing page? https://twitter.now/

"We are Twitter, reclaimed. Operation Bluebird, Inc. picked up the name X Corp. walked away from and is rebuilding it on trust, in your browser at twitter.now. We are not X, and we are not affiliated with X Corp."

"Not a price. Early Access and a stake." Not X but Y

And goes on, I'm staying the hell away from this vibe-coded nightmare.

Comment · 2 months ago
"It simply does not add up that someone who lacks purely technical skills in an area can make good decisions."

Let me introduce you the the executive team of every company.

Comment · 2 months ago
So whats the alternative? I'm thinking I'll migrate my company to something self hosted, but what about personal projects?
Comment · 2 months ago
Its all vibes man
Comment · 2 months ago
This is amazing I been wanting something like for a long time. I'm also curious whats up with the no VC logo at the bottom of the page?
Comment · 2 months ago
Comment · 2 months ago
"The Atmosphere Is Moving Faster Than Anything Evolved to Track It"

Yup AI Slop

Comment · 3 months ago
Looks pretty good, I like that they are highlighting the potential uses for elixir.

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton