_1tem

Security conscious domain. We do automations on behalf of our clients and they don't want credentials stored. "Handling" 2FA automatically is completely unacceptable, it breaks the entire point of the 2FA security model. Besides, login sometimes involves out-of-band 2FA methods including phone number.

What about hybrid automations or human-in-the-loop flows? We have automations where the human starts by logging in, then hands over to the agent. Some parts may even be Puppeteer automated. This also means the session may be long running, typically for months at a time and the agent needs to notify the human again if they get logged out. None of the existing browser automation platforms I have tried make this easy or cost effective, so we are currently trying to build our own. Would love to consider Simplex if this is solved.

Sounds awesome but can I use it like a regular database for my Laravel/Rails/Django app?

And not just any video data, they specifically mentioned screen recordings for agentic computer uses. A very specific kind of video. My guess is they have a partnership with someone like Rewind.ai

I wonder if this would enable the truly "serverless" application I've been thinking about. Imagine shipping a whole Rails/Laravel/Wordpress app to the user to be run in their browser with sqlite. Technically you would only need a CDN to distribute the app.

It's not clear to me what a "container" and "pairing" is in this context. What if my application is not dockerized? Can Claude Code execute tests by itself in the context of the container when not paired? This requires all the dependencies, database, etc. - do they all share the same database? Running full containerized applications with many versions of Postgres at the same time sounds very heavy for a dev laptop. But if you don't isolate the database across parallel agents that means you have to worry about database conflicts, which sounds nasty.

In general I'm not even sure if the extra cognitive overload of agent multiplexing would save me time in the long run. I think I still prefer to work on one task at a time for the sake of quality and thoroughness.

However the feature I was most looking forward to is a mobile integration to check the agent status while away from keyboard, from my phone.

People who want the Good Stuff in life acquire taste and expertise or rely on the opinion of trusted people who have taste and expertise. It's always been like this. Otherwise acquiring the Good Stuff is a matter of random chance.

This is not just for LLM code. This is for any code that is written by anyone except yourself. A new engineer at Google, for example, cannot hit the ground running and make significant changes to the Google algorithm without months of "comprehension debt" to pay off.

However, code that is well-designed by humans tends to be easier to understand than LLM spaghetti.

I, as an engineer, absolutely support keeping private APIs private in order to preserve the performance of products. This is a good policy decision and a good engineering decision. Allowing others to build products on top of private APIs creates false user expectations. The user doesn't know the difference between a public or private API, they just expect their devices to work properly. If you allow a private API to be abused, then third parties may create crappy products (such as fake AirPods) which will ruin the battery life and security of Apple devices. The user doesn't know who to blame, Apple or the third party. It is absolutely within Apple's rights to protect their private APIs from misuse in order to preserve security and performance of their products for all users. The user is free to choose a different brand if they want interoperability. I, as an Apple user, want security and performance, not interoperability. If I wanted interoperability I would choose Android. The EU has no right to force Apple to become crappier like Android.

In order to build a high performance product you have to control all of the API surfaces. Apple devices battery life would suffer if any shoddily written third party device could operate with crappy drivers for it.

It’s about the fact that opening your API limits your own engineering capabilities. A device with an open API has a different security and performance profile.

So Apple has to sabotage their own devices performance and security to let other people use it. The EU has no business in this.

If Apple provided public APIs for their products their own mobile devices battery life and security would be worse due to crappy third party integrations. Apple achieves high performance experiences precisely because that is a ENGINEERING REQUIREMENT to build high quality products.

The 25% number you quoted is for the entire European market including Russia and the Middle East not just the EU. Actual EU market is around 7%.

Anyway the point is, Apple lost 0 market share even after they pulled out.

The Russian market is not tiny, it is literally the largest economy in Europe with the highest GDP per capita for 2025 when adjusted for PPP. It is also has the highest population in Europe.

If they can pull out of Russia and lose nothing why can’t they pull out of the EU?

Despite Apple’s exit from Russia after the 2022 invasion, StatCounter data shows iOS share actually rose from ~27.5% in Aug 2021 to ~32% in 2025 in Russia.

If Apple left the EU they would lose nothing.

If we want to see what the actual impact on Apple’s market share would be if they pulled out of the EU market, consider Russia as a case study. Since Apple pulled out of Russia in 2022, the market share of iOS devices has remained steady and even slightly increased. Russia is comparatively poor, Apple would lose nothing if they pulled out of the EU because consumers would simply travel to other countries to buy.

Here’s a comparison of iOS share in Russia (mobile OS) between August 2025 vs August 2021, using StatCounter data:

• August 2025: iOS ~ 31.97 %

• August 2021: In 2021, the iOS share in Russia was about 27.52 % (for mobile OS) per StatCounter’s data.