I used to do it but now I only review code organization and structure.

Mostly I define the product behavior I want and discuss the technical implementation beforehand.

Once that's in a spec I let it build in my durable implementation engine. Automated verification, review rounds from different agent types across different models, issues triaged into meaningful clusters and flagged for false positives before finally being fixed.

That loops until the code is up to spec.

There is no special secret, LLMs need repetition and multiple passes.

They'll one shot bigger and bigger things in their default harnesses but they'll be filled with issues no matter what.

That's crazy

I'd upgrade my Shield Pro to something else but there's nothing to upgrade to that doesn't have some compromise on formats or capabilities.

Got a bunch of Daytona credits which are close to useless given their restrictions until you hit tier 3.

You seem like you're actually slightly cheaper than https://boat.dev/ but with less storage? That's the closest to your pricing unless I'm mistaken. Any other insight as to differences between you two?

Also I'll add you to a list I've been maintaining of sandbox providers, congrats on the launch!

https://engine.build/lab/agent-sandboxes

But auto renewals not working at times for some reason, credit cards not being saved, prices rising vs competitors.

Just switched everything to CloudFlare since I'm always pointing to it anyways and use a lot of their services.

The switch to Vaultwarden was insanely easy.

Built my own in a day that blows them out of the water. You can quite literally pick any sufficiently good local model or API provider and combine it with Cerebras for cheap and very fast AI post processing and formatting.

Anthropic isn't even close to being this useful.

I also don't care much about individual conversations with agents as the main way of managing work, my personal dashboard shows projects (and subdivided by work trees underneath) but I have an orchestrator agent who is the one I talk to 90% of the time and who is responsible for summarizing information from various project manager agents and only update tiles in the dashboard when I actually need to do something, in the form of questions I can answer directly.

Over time as I add more preferences for how to work, the orchestrator sends fewer and fewer requests to me that it can handle itself.

I was coding already before LLMs but nowadays any little idea for something useful can be done on 15 minutes.

From your profile, you're a retired neurosurgical anesthesiologist so I'm not sure how much you tinker with code but if you've got a ChatGPT subscription you can just ask ChatGPT to build you useful little utilities and websites like this now.

It will publish it for you either privately or publicly under your own subdomain e.g. something.bookofjoe.chatgpt.site

Could be fun for you to try for anything you can think of.

https://what-are-you-reading-september-2026.aljosa-asanovic....

6.1 will probably perform even better.

Opus 5.5 is better in that regard than Fable 5.1 except for some bugs with more complexity which it was uniquely suited to.

Makes perfect sense.

(If you're an idiot)