cjonas

i did this with opus5.5 this year and it caught an 8000$ mistake. Obviously i should find a new CPA, but i have found it incredibly helpful. One downside is that it will likely also flag a bunch of "by the book" libilities to how your filing that I'm pretty sure most CPA would let slide.

My first experience in antigravity was extremely disappointing.

I asked for a QUICK healthcheck on the sync status of a bigquery integration. It literally spun for 25m before I killed it.

The task, which is CLEARLY outlined in the agent readmes just involved using the gcloud cli to initiate auth, then just run a few queries to check logs and timestamps.

Even 3.6 flash can get this right (most the time).

I wanted to do this with a google home hub so i can run my own fully local "home assisant", but claude rejected it... instead i ordered a used lenovo thinkview device which appetently has a community around custom bootloading...

But it does make you think there might be a real market for unlocked devices like this now that it is trivial to build your own software to do exactly what you want.

I think that sould be obvious without even looking at the code... Like what artcraft did with adobe, its now viable to offer apps completely free by leveraging 100% ai. You could argue the code quality doesnt matter much as long as it works (for local apps security is less a concern).

Also, when building similar style apps, I often don't even start git until the app is in a stable state. The earliest code is often worthless and the overall direction not stable.

i forked handy and built 2 modes. One for dictation and one for quick agent actions that uses the pi harness with local models and jev style classifiers so it can read screens and interact with elements (via accessiblity tree, screenshots, os scripts, bash, mcp, etc). Its nice because in agent mode you can just give it instructions on how to respond (like dictation that can read the context of the current page or input). I've been pretty happy with it so far and debating if i should release it. Just not sure if the world needs yet another vibe coded agent system.

Despite whatever else happens with AI, I am excited about the near future where we have an open source version of basically every app. I hope this is the beginning of the end of the subscription model. Keep it up!

Once everyone operates as you do, then websites will have little value. Just publish a .md file or API if you have interactivity.

This thing is obviously fully vibe coded... Click the "what's coming next" and under the healthcare section the marketing screenshot contains broken images... Like no one ever looked at this before deploying it?

Where does the "goal" come from? Is it just known based on different positions in the game? Otherwise it seems like it would require text generation.

Is it likely not just distilled from one of the flagship models? That sounds like an afternoon of work and a few thousand dollars in tokens?

I agree there is probably some impact on output quality when doing this (Claude seems to embedded key signals all over the writing, which is much or what makes it horrible to read). However this does work well as a "second pass" over technical docs or designs along the way

Seems like a lot of this knowledge was encoded into the blog post. I wonder if given this post and access to a planet scale instance to compare with, how close an agentic agent could get.