Oscalemor

or have another agent review it, right?

I love this type of benchmarking, especially when it to me was counter intuitive that astra missed it. Got a couple of nice prompts and things to think about for my own tax making moving to a new country, but honestly hard for most people to review if they got the optimal output from taxes.

Since it's really a proving game, you can optimise and take more risks with declaration, and some of it has to be proven in an audit. Tried getting data on the risks, how old cases were ruled but ultimately by accountant gave a fairly different view compared to Opus 5 at the time.

Yeah that's the vision for sure. I'll check it out, haven't been able to fully commit to an complete autonomous system

Tried to attribute it by going through the prompts and history and pairing it with the commits got it down to 18% of unknown! Which is not perfect, but hey!

Coming from reviewing mainly MD files, or text output in the terminals I was positively surprised when Cursor Cloud used its VM to make nice screen shots and recordings so you could see the new feature live before deploying.

But doesn't really solve the option A option B review types

I'm having a hard time differentiating all these different techniques, would love to see a broader comparison

Congratulations to the OpenAI team! It's nice that we start looking not only at input-output cost and token maxxing and also begin to look at how it actually completes the task at hand.

Diagrams on this page is a real struggle to read though

Radical transparency, I approve.

Would love for my agent to chime in 'nothing new on my end'

on a serious note, the classic enterprise stack of Gemini and Jira really has put me in a position of questioning the standup. Kept mostly for vibes and routines I would say.

Eager to hear if anyone else have done similar results, would love to build out a real study on this and the actual effects!

Picture myself a room of complete darkness, the lonely duck.