HNHacker News
TopNewBestAskShowJobs

pietz

391 karma · joined December 2, 2016

submissionscomments
pietz··on Datasette Apps: Host custom HTML applications inside Datasette
Hey Simon,

although I'm coming from a different starting point, it seems like some of our thoughts have aligned. I'm building https://caipi.ai/ as a workspace for agents to build simple data driven apps. The agent edits through MCP and the user gets an interactive app in the browser.

If you're interested picking each others brains around this topic, I'd be psyched to have a chat. gh:pietz.

pietz··on Feds freaked over Fable 5 after 'fix this code', not jailbreak, say researchers
Hats off to them for using GPT-2 to design their website.
pietz··on SpaceX to buy Cursor for $60B
It's a crazy number especially since Cursor feels kinda dead. Few thoughts from the other side:

- xAI needs the coding related data to compete with Claude Code and Codex

- Recent progress with Composer 2.5 seems promising given the size

- The may get a comeback on the smaller than Enterprise battle field now that the other two got so expensive

- The way that Elon set up this entire process was quite genius. They locked in this option before, and now after the gains through the IPO, it feels almost like a discount, lol

pietz··on Claude Fable 5
> On June 23, we’ll remove Fable 5 from those plans. Using it after that will require usage credits.

We've entered the phase where only companies will be able to afford state-of-the-art models.

pietz··on Claude Opus 4.8
1. Benchmarks saturate 2. They select the most impressive improvments
pietz··on From Supabase to Clerk to Better Auth
This comment is more ridiculous than ever in 2026.
pietz··on GPT-5.5
OpenAI has been very generous with limit resets. Please don't turn this into a weird expectation to happen whenever something unrelated happens. It would piss me off if I were in their place and I really don't want them to stop.
pietz··on 30 Days Running ChatGPT Plus, Claude Pro, and Google AI Pro in Parallel
> Claude Opus 4.7 ($20 Pro tier) is the new coding king.

That's ridiculous. If you use Opus 4.7 for coding on the $20 tier, you'll hit the session limit within 20 minutes.

pietz··on Cal.com is going closed source
This conclusion makes more sense to me, but maybe I'm too naive.

The media momentum of this threat really came with Mythos, which was like 2 or 3 weeks ago? That seems like a fairly short time to pivot your core principles like that. It sounds to me like they wanted to do this for other business related reasons, but now found an excuse they can sell to the public.

(I might be very wrong here)

pietz··on Trivy under attack again: Widespread GitHub Actions tag compromise secrets
So by wanting to improve the security of my application, I ended up lowering the security of my application? Nice.
pietz··on Nano Banana 2: Google's latest AI image generation model
I'm officially done with the Nano Banana name. It was fun, but can we go back just calling it Gemini Image?
pietz··on Voxtral Transcribe 2
Do we know if this is better than Nvidia Parakeet V3? That has been my go-to model locally and it's hard to imagine there's something even better.
pietz··on Claude Code is suddenly everywhere inside Microsoft
Why not just use it in the terminal? That's literally what it was built for.
pietz··on AGENTS.md outperforms skills in our agent evals
Isn't it obvious that an agent will do better if he internalizes the knowledge on something instead of having the option to request it?

Skills are new. Models haven't been trained on them yet. Give it 2 months.

pietz··on Consent-O-Matic
This idea/execution isn't new right? Can someone explain what makes this different/better? Is this the ublock Origin of cookie banner hiders?
pietz··on Anthropic blocks third-party use of Claude Code subscriptions
Honest question: Why would I use Claude with OpenCode if I have a Claude Max subscription? Why not Claude Code?
pietz··on LMArena is a cancer on AI
Uhm, yes that's why you rely on LMArena (core) results only to judge the answering style and structure. I thought this was common knowledge.
pietz··on Show HN: OfferGridAI – side-by-side comparison of real estate offers from PDFs
2006: "This meeting could have been an e-mail"

2026: "This app could have been a prompt"

pietz··on We removed 80% of our agent's tools
Hold up, so d0 is just Claude with access to the terminal?

Sounds like they just discovered that they don't have a product.

pietz··on Codex is a Slytherin, Claude is a Hufflepuff
Weirdly, I find a higher signal to noise in this analogy than looking at benchmarks these days.

If you let your inner fanboy rest for a moment you realize Gemini 3, Claude Opus 4.5, and GPT 5.2 are all amazing. If two of them disappeared tomorrow, my AI assisted productiveness wouldn't change.

The 3% difference on benchmark X doesn't mean anything anymore. It's probably more helpful to compare them on character traits instead of numbers.

My one word to describe Claude would be "pleasant". It's just so nice to communicate with. GPT/Codex would be the "thorough". It finds and thinks of stuff the others don't. For Gemini 3, the jury is still out. It might be the smart kid on the block that's still a bit rough around the edges, but given that it's a preview things might change soon.

pietz··on Show HN: RenderCV – Open-source CV/resume generator, YAML to PDF
I hear you. This boils town to personal opinion. I would have preferred to use an existing standard than introducing yet another one. The custom sections aren't something I've ever seen or needed anyway.
pietz··on Show HN: RenderCV – Open-source CV/resume generator, YAML to PDF
A bit of a missed opportunity not to use the JSON Resume schema for this.

https://jsonresume.org/schema

pietz··on GPT-5.2-Codex
That's a fair point and yet I deeply believe Codex is better here. After finishing a big task, I used two fresh instances of Claude and Codex to review it. Codex finds more issues in ~9 out of 10 cases.

While I prefer the way Claude speaks and writes code, there is no doubt that whatever Codex does is more thorough.

pietz··on Gemini 3 Flash: Frontier intelligence built for speed
Interesting. Flash suggests more power to me than Mini. I never use gpt-5-mini in the UI whereas Flash appears to be just as good as Pro just a lot faster.
pietz··on I ported JustHTML from Python to JavaScript with Codex CLI and GPT-5.2 in hours
Now do the same with Rust, build a Python wrapper and we went full circle :)
pietz··on 7 Years, 2 Rebuilds, 40K+ Stars: Milvus Recap and Roadmap
People buy F150s because they find them cool and not because they actually need the space. Your Corolla could make deliveries around town in roughly the same time, while being cheaper and easier compared to introducing a new expensive car. In situations you need more space (which most of us won't), you can add a trailer instead.

Interesting, I guess we're on the same page ;)

pietz··on 7 Years, 2 Rebuilds, 40K+ Stars: Milvus Recap and Roadmap
You start by underselling what can be done with Postgres and then follow up with the upper end of requirements that most projects won't need. My argument is exactly what you conveniently left out: The big bulk between the two.
pietz··on 7 Years, 2 Rebuilds, 40K+ Stars: Milvus Recap and Roadmap
It seems to me like many developers are moving from native vector stores back to Postgres with pgvector. It's easier to integrate, cheaper to set up, familiar to work with, oftentimes faster, while not hitting any limitations for most projects. My vector store recommendation boils down to: "Use pgvector unless you have specific reason not to".

Does this still hold?

pietz··on GPT-5.2
They are definitely ahead in multi modality and I'd argue they have been for a long time. Their image understanding was already great, when their core LLM was still terrible.
pietz··on [dead]
It means half of the internet isn't (wasn't) working.
← PreviousPage 3 of 6Next →