swe_dima

10 ms·

1.
by swe_dima 11d ago
According to the metrics the "fast" mode is not any faster...
2.
by swe_dima 24d ago
overly high demand can often put a system in an unstable state
3.
by swe_dima 27d ago
I don't understand why they don't update the status page. Been going on for 1+ hour. Is it counted as "uptime"?
4.
by swe_dima 1mo ago
my prompt (doesn't help): > Go easy on the comments, only add comments if there's a big gotcha that is not clear from the code itself, or if something in another place is going to cause a side effect. Code should be self-docume
5.
by swe_dima 1mo ago
My feeling that as it becomes a better coder it becomes a worse communicater. It's overfitting for coding benchmarks, while communication style is harder to quantify during training. And no matter how often I tell it to stop adding com
6.
by swe_dima 1mo ago
It's why I love React Native and over the air updates. So I don't have to deal with this roulette.
7.
by swe_dima 2mo ago
In my experience SVGs are still too hard for LLMs. I gave Fable a jpeg and asked to draw an SVG, using a loop that renders the SVG into an image so Fable can inspect it. Results looked like drawing of a 5 year old.
8.
by swe_dima 2mo ago
I think by default they check out a new connection when obtained a job and then release it. In comparison with some async languages like JS where a connection is only checked when a query is about to be executed
9.
by swe_dima 2mo ago
My concern is number of database connections. In the example it's 100 fibers per worker, at that rate you are going to exhaust db connections sooner. Happy to be wrong.
10.
Octane – compiled and faster React from creator of Inferno(octanejs.dev)
3 points by swe_dima 2mo ago|1 comments
11.
by swe_dima 2mo ago
I also love them - makes spotting AI written content easier. People debated implementing complex algorithms how to watermark AI content - but luckily AI just settled on it by itself. (sorry for a bit of gaslighting)
12.
by swe_dima 2mo ago
Gemini flash lite family of models currently has the best ratio for price/speed/intelligence for understanding images, no real alternative AFAIK
13.
by swe_dima 2mo ago
It's scary relying on Google's models. I have a very price sensitive workload that used to run on flash 2.5 lite - it's deprecated now. The replacement 3.1 flash lite is a lot more expensive, but now also has a sunset date. 3
14.
by swe_dima 2mo ago
About the same, using cloudflare. For some very bizarre reason Vertex has a very low non-adjustable quota.
15.
by swe_dima 2mo ago
the writing is on the wall for it, I have switched to gemma-4-26b-a4b. At least in benchmarks, it scores higher and is faster.
16.
by swe_dima 2mo ago
"You were the chosen one! You were supposed to destroy them, not join them!"
17.
by swe_dima 3mo ago
if I travel and expect to be working I now bring a CO2 meter
18.
by swe_dima 3mo ago
If agent is asking a question it means it stumbled on something tricky and often an important architectural decision has to be made. Even if I am staring at the screen at the time, it's often impossible to study code and make a judgeme
19.
by swe_dima 3mo ago
Is it possible to use their subscription pricing with Opencode?
20.
by swe_dima 3mo ago
many years ago when Hanami was just getting popular I remember doing benchmarks against Rails when it comes to SQL and was unpleasantly surprised when Rails' ActiveRecord ended up being much faster, despite "speed" being adve
21.
by swe_dima 3mo ago
Not sure what niche it's going to occupy: too expensive for it's intelligence category.
22.
by swe_dima 3mo ago
Pleasantly surprised that it costs as GPT 5.5, thank god for the competition.
23.
by swe_dima 3mo ago
I found Ruby LLM to be surprisingly good - in terms of usability it's close to Vercel's AI framework. It tries to strike a balance between working out of the box and being flexible... which has its challenges, still nice overall.
24.
by swe_dima 3mo ago
Love Vite, but always felt sorry for them because it was not clear how they can make money, the whole VoidZero thing felt like a stretch. It's one of those things that always stopped me from building cool tools - you have to make a liv
25.
by swe_dima 4mo ago
I still don't understand how people use git worktrees with Docker. You need a full database and etc. For me it's simpler to have multiple checkouts.
26.
by swe_dima 4mo ago
the video on this post looked like a Tiktok - everything is jumping, I could barely grasp what was going on
27.
by swe_dima 4mo ago
You may remember the argument that you can build an AI app and it continues to improve as models improve and costs go down? Well, looking at OpenAI / Google / Anthropic we see crazy cost increases, such that it might invalidate yo
28.
by swe_dima 4mo ago
Flash family but costs like a Pro. $9 vs $12 for output.
29.
by swe_dima 4mo ago
Is it basically about giving as much data as possible to insurance companies?
30.
by swe_dima 7mo ago
fair enough, thanks for elaborating

More ›