Hacker News: swe_dima

New comment by swe_dima in "Why does Opus 5 feel worse to work with?"

swe_dima — Fri, 14 Aug 2026 16:25:02 +0000

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-documenting. When in doubt, don't add a comment at all. If you do have to add a comment, make it short and on point. Comments should show history of code changes or functionality, only comment on the current state (or not at all).

New comment by swe_dima in "Why does Opus 5 feel worse to work with?"

swe_dima — Fri, 14 Aug 2026 12:31:59 +0000

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 comments it just can't help itself.

New comment by swe_dima in "App Store Rejection of the Week: Dark Hours"

swe_dima — Fri, 07 Aug 2026 20:16:18 +0000

It's why I love React Native and over the air updates. So I don't have to deal with this roulette.

New comment by swe_dima in "Karpathy’s Pelican"

swe_dima — Sun, 02 Aug 2026 20:26:40 +0000

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.

New comment by swe_dima in "Solid Queue 1.6.0 now supports fiber workers"

swe_dima — Sat, 01 Aug 2026 13:29:44 +0000

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

New comment by swe_dima in "Solid Queue 1.6.0 now supports fiber workers"

swe_dima — Sat, 01 Aug 2026 11:50:24 +0000

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.

Octane – compiled and faster React from creator of Inferno

swe_dima — Mon, 27 Jul 2026 20:27:20 +0000

Article URL: https://octanejs.dev

Comments URL: https://news.ycombinator.com/item?id=49075107

Points: 3

# Comments: 1

New comment by swe_dima in "Em dashes are fucking amazing"

swe_dima — Fri, 24 Jul 2026 13:55:42 +0000

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)

New comment by swe_dima in "Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber"

swe_dima — Tue, 21 Jul 2026 19:42:42 +0000

Gemini flash lite family of models currently has the best ratio for price/speed/intelligence for understanding images, no real alternative AFAIK

New comment by swe_dima in "Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber"

swe_dima — Tue, 21 Jul 2026 16:04:36 +0000

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.

3.5 flash lite is even more expensive.

So the price is rising and you have no choice but to keep paying more and more.

New comment by swe_dima in "Please don't discontinue Gemini 2.5 Flash"

swe_dima — Sat, 11 Jul 2026 21:37:53 +0000

About the same, using cloudflare. For some very bizarre reason Vertex has a very low non-adjustable quota.

New comment by swe_dima in "Please don't discontinue Gemini 2.5 Flash"

swe_dima — Fri, 10 Jul 2026 21:31:53 +0000

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.

New comment by swe_dima in "Beijing is looking at curbing overseas access to China's top AI models"

swe_dima — Tue, 07 Jul 2026 17:07:31 +0000

"You were the chosen one! You were supposed to destroy them, not join them!"

New comment by swe_dima in "The bottleneck might be the air in the room"

swe_dima — Sat, 04 Jul 2026 09:34:44 +0000

if I travel and expect to be working I now bring a CO2 meter

New comment by swe_dima in "Claude's AskUserQuestion: "No response after 60s – continued without an answer""

swe_dima — Thu, 02 Jul 2026 19:59:22 +0000

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 judgement call that quickly.

My point being, even for opt-in it can be too short - highlights how little consideration they put into it, just releasing slop in production.

New comment by swe_dima in "ZCode"

swe_dima — Wed, 01 Jul 2026 20:27:22 +0000

Is it possible to use their subscription pricing with Opencode?

New comment by swe_dima in "Hanami 3.0: In Full Bloom"

swe_dima — Wed, 01 Jul 2026 20:22:25 +0000

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 advertised as one of the advantages :-(

New comment by swe_dima in "Claude Sonnet 5"

swe_dima — Tue, 30 Jun 2026 19:44:43 +0000

Not sure what niche it's going to occupy: too expensive for it's intelligence category.

New comment by swe_dima in "Previewing GPT‑5.6 Sol: a next-generation model"

swe_dima — Fri, 26 Jun 2026 18:11:41 +0000

Pleasantly surprised that it costs as GPT 5.5, thank god for the competition.

New comment by swe_dima in "RubyLLM: A Ruby framework for all major AI providers"

swe_dima — Wed, 24 Jun 2026 15:11:26 +0000

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.

One big real-life pain I experienced is that caches don't always work, e.g. for xAI, since it only supports completions API and thought signatures are returned wrong.