Hacker News: dkersten

New comment by dkersten in "Pro Max 5x quota exhausted in 1.5 hours despite moderate usage"

dkersten — Mon, 13 Apr 2026 07:24:00 +0000

Yet he vibe slops the code that the customer has to use.

New comment by dkersten in "Pro Max 5x quota exhausted in 1.5 hours despite moderate usage"

dkersten — Sun, 12 Apr 2026 15:41:37 +0000

Eh you say that every time and yet it keeps happening.

New comment by dkersten in "Project Glasswing: Securing critical software for the AI era"

dkersten — Wed, 08 Apr 2026 12:45:45 +0000

And every single time what they release is underwhelming.

Remember how Sam spent like a year talking about how scary close GPT-5 was to AGI and then when it did finally come out... it was kinda meh.

New comment by dkersten in "Components of a Coding Agent"

dkersten — Sun, 05 Apr 2026 09:49:39 +0000

I’m building something similar. It’s not public yet because it’s still early and I’m still working on exactly what it is supposed to be.

But the idea is similar in that I start with a spec and feed the LLM context that is a projection of the code and spec, rather than a conversation. The context is specific to the specific workflow stage (eg planning needs different context to implementing) and it doesn’t accumulate and grow (at least, the growth is limited and based on the tool call loop, not on the entire process).

My main goals are more focused context, no drift due to accumulated context, and code-driven workflows (the LLM doesn’t control the RPI workflow, my code does).

It’s built as a workflow engine so that it’s easy for me to experiment with and iterate on ideas.

I like your idea of using TOML as the artifact that flow between workflow stages, I will see if that’s something that might be useful for me too!

New comment by dkersten in "AI overly affirms users asking for personal advice"

dkersten — Sun, 29 Mar 2026 09:57:02 +0000

I was typing quickly on my phone. I meant “quite”, “I find Kimi quite good”

New comment by dkersten in "AI overly affirms users asking for personal advice"

dkersten — Sat, 28 Mar 2026 15:34:01 +0000

I find Kimi white good if you ask it for critical feedback.

It’s BRUTAL but offers solutions.

New comment by dkersten in "$500 GPU outperforms Claude Sonnet on coding benchmarks"

dkersten — Sat, 28 Mar 2026 09:45:26 +0000

I’ve never had any problems with MiniMax. I wouldn’t call the speed fast exactly, but it’s faster than GLM and seems similar to Opus.

It’s been fast enough that I’ve been using it as my main model (M2.7 and before that, M2.5). Opus still does better at tasks, but MiniMax is so much cheaper. I’ve used their cheaper plan and I’ve never been rate limited.

New comment by dkersten in "$500 GPU outperforms Claude Sonnet on coding benchmarks"

dkersten — Sat, 28 Mar 2026 09:42:30 +0000

I’ve also never hit the MiniMax limits and M2.7 is pretty good.

Not as good as Opus, but substantially cheaper!

New comment by dkersten in "Wine 11 rewrites how Linux runs Windows games at kernel with massive speed gains"

dkersten — Wed, 25 Mar 2026 08:12:11 +0000

They could brand it as “New Windows”

New comment by dkersten in "Ask HN: Are MiniMax Models Scams?"

dkersten — Mon, 23 Mar 2026 21:01:07 +0000

> Did anyone actually use a MiniMax model and it actually worked?

Works well for me in Kilo Code. Its not as good as Opus, but its substantially cheaper and gets the job done. I don't have any problems with it.

New comment by dkersten in "How I write software with LLMs"

dkersten — Tue, 17 Mar 2026 16:24:31 +0000

I do. I tend to follow a strict Research, Plan, Implement workflow. It does greatly help, but it doesn’t eliminate all problems.

New comment by dkersten in "How I write software with LLMs"

dkersten — Mon, 16 Mar 2026 15:59:34 +0000

> At least the LLM will only take 5 minutes to tell you they don't know what to do.

In my experience, the LLM will happily try the wrong thing over and over for hours. It rarely will say it doesn’t know.

New comment by dkersten in "Please Do Not A/B Test My Workflow"

dkersten — Sat, 14 Mar 2026 17:39:35 +0000

I don’t think they’re even saving much on vibe coding it, given how many tokens they claim they’re using. I know the token cost to them is much, much lower than the token cost to us, but it still has a cost in terms of gpus running.

Plus it’s not something we can replicate since we don’t have access to infinite tokens, so it’s not even a good dogfooding case study.

New comment by dkersten in "Please Do Not A/B Test My Workflow"

dkersten — Sat, 14 Mar 2026 17:37:23 +0000

I get the value of dogfooding, but I feel that in this case, a solid trustworthy foundation is much more important than dogfooding.

New comment by dkersten in "Please do not A/B test my workflow"

dkersten — Sat, 14 Mar 2026 12:06:00 +0000

Anthropic have done a lot of things that would give me pause about trusting them in a professional context. They are anything but transparent, for example about the quota limits. Their vibe coded Claude code cli releases are a buggy mess too. Also the model quality inconsistency: before a new model release, there’s a week or two where their previous model is garbage.

A/B testing is fine in itself, you need to learn about improvements somehow, but this seems to be A/B testing cost saving optimisations rather than to provide the user with a better experience. Less transparency is rarely good.

This isn’t what I want from a professional tool. For business, we need consistency and reliability.

New comment by dkersten in "We should revisit literate programming in the agent era"

dkersten — Mon, 09 Mar 2026 11:40:23 +0000

Code alone can never describe intent or rationale.

New comment by dkersten in "The L in "LLM" Stands for Lying"

dkersten — Thu, 05 Mar 2026 16:35:26 +0000

Almost every 3D game in the past 20 years uses procedural foliage generation (eg SpeedTree and similar). Many use procedural terrain painting. Many use tools like Houdini.

So procedural generation is extremely prevalent in most AAA games and has been for a long time.

New comment by dkersten in "iPhone 17e"

dkersten — Tue, 03 Mar 2026 08:04:24 +0000

I loved the size of my iPhone 6, and very iPhone that I’ve used after that has been too big.

New comment by dkersten in "IDF killed Gaza aid workers at point blank range in 2025 massacre: Report"

dkersten — Sun, 01 Mar 2026 11:56:51 +0000

And the IDF are terrorists.

New comment by dkersten in "Minions: Stripe’s one-shot, end-to-end coding agents"

dkersten — Sun, 22 Feb 2026 15:29:04 +0000

I can find it off putting regardless. Especially since I’m not the person who released it under MIT license.