Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
antupis
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
8 ms
·
61.
▲
by
antupis
2y ago
Have you tested tool calling capabilities of these cheaper models?
62.
▲
by
antupis
2y ago
This has a strong vibe of the past: "In the early 1980s AT&T asked McKinsey to estimate how many cellular phones would be in use in the world at the turn of the century. The consultancy noted all the problems with the new devices—t
63.
▲
by
antupis
2y ago
Yup, there is zero chance that the market is too small. The only issues are whether a company can the technology to work and obtain regulatory approval.
64.
▲
by
antupis
2y ago
Google has very research division but productising those inventions is the problem.
65.
▲
by
antupis
2y ago
Yeah this working with knowledgeable customer is like magic.
66.
▲
by
antupis
2y ago
Not true theirs is plenty of mediocre players that become great coaches. I really cannot see why IT would be different.
67.
▲
by
antupis
2y ago
A big part is that you are not working for man but yourself.
68.
▲
by
antupis
2y ago
> Managers need to pay close attention to where they spend their energy (=time). For example, the CEO should not be spending time picking items for the company snack cupboard. Last startup where I was working, I knew that the company was
69.
▲
by
antupis
2y ago
When you ship it is arts and crafts which usually leans heavily to a small group of people Then if you want it to be commercial succesfully it needs some manufacturing practices because otherwise, you end up with legacy software.
70.
▲
by
antupis
2y ago
Mastodon would be perfect for these but there is no critical mass at least yet.
71.
▲
by
antupis
2y ago
If end-users do like it upper managment notices it but it will take time, generally you iterate faster against upper managnent. Same kinda applies when end users dont like that stuff so you cannot lean too heavily to opticks game.
72.
▲
by
antupis
2y ago
And it is working generally very well if you compare GCP and Azure. DX is much better at GCP but Azure does stuff that companies care much better and is 2x bigger.
73.
▲
by
antupis
2y ago
Yeah, luckily southern Sweden has about 1-2 hours more daylight.
74.
▲
by
antupis
2y ago
It is kinda hard when there is no snow so it is really pitch black. Basically October-November sucks and then it gets easier.
75.
▲
by
antupis
2y ago
We kind of see same shift happening in real-time with The Beatles at the moment
76.
▲
by
antupis
2y ago
Stupid question but is this wise from a business standpoint? let's say child pornography or some national security stuff slips into training data.
77.
▲
by
antupis
2y ago
It is so interesting that those are always so country-specific here in Finland it is Medicine and industrial management.
78.
▲
by
antupis
2y ago
Generally, these cloak-and-dagger companies just vanish especially industry where is fierce competition because cloak-and-dagger stuff takes just too much energy from the north start. I would not be surprised if 2025 GPT-5 still is not out
79.
▲
by
antupis
2y ago
My ex colleague is still fuming after several years about case where professional manager decided to program parser that we decided was 8 story points. If this manager would have been founder it would probably have been normal stuff about
80.
▲
by
antupis
2y ago
Thing is that LLM generated word salad is destroying Google search.
81.
▲
by
antupis
2y ago
Yup that is how I would do it. Water sample returns to labs > water mining > platinum/gold/etc.
82.
▲
by
antupis
2y ago
> As everything, now AI is losing hype, but somehow (in my bubble) seems like engineers are still high on it. But I also see that this will distill further the set of people who I look up to and want to collaborate with, because e of tha
83.
▲
by
antupis
2y ago
Is there a site that has hand-curated playlists I would love that let's say if I want to listen to Korean pop from the 90s or Minimal Techno from the 00s.
84.
▲
by
antupis
2y ago
Also tenure Glass would have been just different than blue collar Glass we had.
85.
▲
by
antupis
2y ago
But usually, in spaces where you need speed Python is just an orchestrator or glue between pipelines, and actually, calculations are done by db or some c/c++/fortran library.
86.
▲
by
antupis
2y ago
Search is more about data, LLMs are somewhere between those two.
87.
▲
by
antupis
2y ago
Also, look for those goldrush phases where there is lots of money thrown around eg now AI,
88.
▲
by
antupis
2y ago
When you have hundreds ci/cd pipelines written on GitLab and people are using ci/cd pipelines more like CRON migration becomes extremely hard.
89.
▲
by
antupis
2y ago
Well yeah but the problem was that you still needed that runtime, WASM should solve this.
90.
▲
by
antupis
2y ago
Cost is a plus but at least what I see is that getting good response time is even bigger. Something like OpenAI Azure instances are inconsistent and it is far too normal to get a 40sec lag with responses with gpt4-o.
More ›