Those are routing decisions. Model price is simply one field in the decision.

The choices made there determine waiting time, failed runs, review overhead, and how often a developer needs to take back control.

Classifying ten thousand tickets has different requirements from analysing an unfamiliar codebase.

A UI review needs visual input.

A security-sensitive change carries a higher verification cost.

Independent searches can run concurrently.

Available here: github.com/gmoigneu/ski... or through www.skills.sh/gmoigneu/ski...

The right UX is structured entity screens plus an assistant with the tools to act on them. Sometimes I just want a quick search, or to glance at the last records. I don't need an LLM for that.
Rebuilding my setup around this idea.

Two years later, researchers from OpenAI, Anthropic, and Google bypassed every published defense at 90%+ success rates. There is no deterministic fix for prompt injection.

I wrote up what I shipped, what I skipped, and why. Let me know if I missed anything important!

I wrote up the full paper breakdown with a practical workflow here: devcenter.upsun.com/posts/agents...

#codingagents #softwareengineering #aiengineering #claude

The agents weren't ignoring the files. They followed every line too literally, treating each instruction as another constraint.

What worked: start empty, watch the agent fail, add one rule when you see the same mistake twice. Five lines that fix real problems.

Really let you think twice of what you can build that can't be easily replicated. And I guess that's where you need to have more human services, support, governance, sovereignty because the product itself may be just a commodity.

#ai #builder #indiedev #claude #anthropic

Many startups started doing this a few months back. It's insane and scary to see how fast Anthropic and OpenAI can replicate a working business and just kill it. Because like Google 10 years ago, there is no reason to have multiple providers when you can just use one instead.

But transcription quality is comparable, speed might be better on a desktop GPU, and everything stays local. For a free open-source tool it fills a real gap. Full walkthrough with build instructions, GPU setup, and compositor keybindings: guillaume.id/blog/how-i-g...

The accuracy surprised me too. Technical vocab, proper nouns, correct punctuation. Good enough that first drafts need less editing than what I type. Honest take: SuperWhisper has better UX. Visual feedback, mode system, polish. hyprvoice is a daemon you toggle from a hotkey. No GUI, no frills.

Despite the name, it works on any Wayland compositor. I run it on niri. You can point it at cloud APIs or run everything locally with whisper.cpp. With whisper.cpp and the large-v3-turbo model on GPU: sub-100ms transcription latency. Even after a full minute of speech. It felt wrong, it was so fast.

Linux options have been rough. Electron wrappers, X11 hacks that break on Wayland, or shipping your audio to the cloud. Nothing survived daily use. Then I found hyprvoice. Go daemon that captures audio via PipeWire, runs it through a transcription backend, and injects text into your focused window.