mikkelam

5 ms·

1.
by mikkelam 5d ago
This is once again RLHF loops. the AI labs are and have been 100% focused on correctness because it is easy to setup and validate. Adding one more function that almost does the same thing as another will not break anything. I think this is
2.
by mikkelam 28d ago
Except horizontal scaling. But a lot of companies are trying to solve that, notably multigres, neki and even pgdog.
3.
by mikkelam 1mo ago
Openrouter is super well positioned in the market. I've always believed in staying model agnostic and being able to jump around providers as their intelligence, cost per intelligence and/or latency issues arise. Openrouter will wi
4.
by mikkelam 1mo ago
Maybe after double digit million death count. Or just a rogue country that's being hit by the crisis more than others
5.
by mikkelam 1mo ago
Definitely not touching this until it's open source
6.
by mikkelam 2mo ago
Zulip is bad
7.
by mikkelam 2mo ago
Super happy to see someone take on slack. We just want a performant chat with simple features. Slack integrations are overrated. Just give me webhooks.
8.
by mikkelam 3mo ago
Would love to see benchmarks on cognition's FrontierCode
9.
by mikkelam 3mo ago
I cannot believe they're sticking to their guns on this website design. It's awful.
10.
by mikkelam 3mo ago
Linear is probably starting to hit enshittification territory. The feature bload is immense and they are trying extremely hard to not be left behind on the AI wave.
11.
by mikkelam 3mo ago
My own experience is far from this. Steering the AI, early and while developing a change, matters sigificantly. Therefore a maintainer is more likely to steer the AI in a direction that is aligned with the codebase.
12.
by mikkelam 4mo ago
The enshittification has begun. Time to move on!
13.
by mikkelam 5mo ago
Just improve what you have GitHub. Stop the AI bloatware. You will lose that race anyway, obviously.
14.
by mikkelam 5mo ago
I've been super happy with my Framework desktop. And since getting that I've been craving replacing my MacBook... This looks super attractive
15.
by mikkelam 6mo ago
Why are we treating LLM evaluation like a vibe check rather than an engineering problem? Most "Model X > Model Y" takes on HN these days (and everywhere) seem based on an hour of unscientific manual prompting. Are we actually r
16.
by mikkelam 7mo ago
For me, GitHub CLI is the prime example of this. This CLI is so incredibly powerful when combined with regular command line tools. Agents know how to use head, tail, jq and so on to only extract the parts it needs. The best selling point of
17.
by mikkelam 9mo ago
They have a very nice comparison in terms of performance and price https://planetscale.com/benchmarks/aurora
18.
by mikkelam 10mo ago
I agree. If you strictly follow the syntax of "Example 1" in JavaScript (calling and awaiting on the same line), the observable output is identical to Python. I suppose the author meant to say that if you first called your async f
19.
by mikkelam 1y ago
Totally. It's a very boring part of this that one would have to contend with. I kinda feel the same way about Facebook. Groups, events, marketplace are amazing for community building. But it's just so hard to compete with Meta.
20.
by mikkelam 1y ago
Safe can mean a lot of things to different people i guess. I would love to incorporate some sort of reputation signal. Perhaps positive reinforcement after people have met? Or just having social links? But yeah, i dont have it all figured o
21.
by mikkelam 1y ago
I think something like the matrix protocol would be better. I would be especially interested in not storing unencrypted user messages. Matrix would be a good choice for this.
22.
by mikkelam 1y ago
I've been thinking a lot about building an open source dating app as a non-profit offering. I have a sense that succesful dating contributes highly to overall human happiness. It should be a public service similar to wikipedia or libr
23.
by mikkelam 1y ago
>While I respect anyone’s decision to spend their days playing pickleball, that life isn’t quite for me—at least not full time. I’m lucky to wake up every day energized to go to work Bit of an unfair comparison though.. Most people dont
24.
by mikkelam 2y ago
How do you know if someone uses Arch Linux? Don't worry they'll tell you ;) All kidding aside, I recently migrated to EndeavourOS, but CachyOS looks dope too
25.
by mikkelam 2y ago
https://www.youtube.com/watch?v=6mmskXXetcg
26.
by mikkelam 2y ago
AI
27.
by mikkelam 2y ago
I have honestly been so excited to try this after listening to several videos of mitchell talking about his work. What a christmas present! A terminal is so dear to us software engineers, and this seems like such a love declaration to the t
28.
by mikkelam 2y ago
I agree on the first part. Having used kagi for about 6 months it often lacks behind on recent things. I find myself automatically adding !g on such queries
29.
by mikkelam 2y ago
at this point they should just start their own Python implementation :-)
30.
by mikkelam 2y ago
The main reason im not concerned about AI-based entertainment is the same reason I watch human chess players. It's not only about technical capabilities. I can't explain fully why though..

More ›