What I see most engineers do is parallelize. Have multiple agents work at the same time. It takes some time getting used to, but I think they are on to something.
There's even an article I read about this the other week, but I can't seem to find it ATM.
Right. The idea here is to kick of 3-8 or something tasks. They finish as you finish writing the next prompt. Then you go and review/test/merge the code from the first task, then another task finishes and you review/test/merge that code.
The challenge is that you have to be working on multiple work streams at once because so far Codex isn't great at not doing work you are doing in another task even if you tell it something like "class X will have a function that returns y"...it will go write that function most times.
I've found it really good for integration work between frontend and backend features where you can iterate on both simultaneously if the code isn't in the same codebase.
Also, for Codex this works best in the web ui because it actually uses branches, opens prs, etc. I think (though could be wrong) that locally with the CLI or IDE extension you might have to manually great git worktrees, etc.
Yeah I try to keep it away from overlapping it's work as much as possible. Using plan mode in claude or just telling codex to build a plan, that is structured in a parallelized way for multiple agents usually helps delegate tasks to be handled at the same time. Typically: app code, infra, and data layer are the main three, but obviously depends on the project.
If I ever find my self just waiting, then it always gives me an opportunity to respond to messages, emails, or update tickets. Won't be long now until the agents are doing that as well...
I can understand how you can parallelize tasks when there is some structure - but how do you get to that initially structured state where each agent isn't recreating the world to conflict with every other agent?
When i use that approach I end up merging one PR, then have to prod the others to fix themselves - resolving conflicts, removing duplicate code, etc - so it ends up slower than just running one agent at a time.
Like i said - maybe this is a problem on a bare repo? But if so, how are people vibe coding from scratch and calling themselves productive? I just don't get it.
I used something similar in the past. Loved it. User wants their data? Bang! Here's the entire database. User deletes their account? `rm username.sql` and you're done. Compliance was never easier.
> In January 2020, BPA was restricted from use in thermal paper, including tickets and receipts, across the EU (3). As a result, another bisphenol, Bisphenol-S (BPS), began to take its place. In fact, an ECHA survey estimated that 61% of all thermal paper would contain BPS as a substitute for BPA, despite concerns of BPS being equally as harmful (6). Fast forward three years and BPS is now recognised as “toxic to reproduction” and a hormone disruptor, and has been added to the EU’s candidate list for Substance of Very High Concern (SVHCs), a common first step on the road to restriction (7).
The mentioned company above (Exacompta) also make some without BPS (they say "sans phenol"). No idea what they use instead, for all I know it could be worse ^^ but I think the made in France is encouraging, we tend to have safer norms than EU which itself tends to have safer norms than the world.
You can see this at play with the support of the Ukraine. Billions are being tossed around but how much 'punch' does that dollar have when it reaches the frontline? Does it even get there at all? Is 10B in American aid even close to 10B in Estonian aid?
Zero provenance of where each aid dollar goes. Does it reach the frontline, who's pocket does it go into? And why can't we track this information (at a high level)?
It's being tracked and there's been government reports checking if the money is being siphoned to corruption and so far not much has been found (except the usual low percentage of defrauding but that happens with anything).
You and I don't have access to it, it's just expected since it's sensitive information during wartime.
Just because your government can break your opsec doesn’t imply it’s not worth doing your best. Maybe you’ll avoid being caught in a lower-cost/mass surveillance system and you might not do anything to attract a more dedicated attack.
Re fingerprint: there’s risks around having your fingerprint lifted from something you touch (eg a glass). It’s a movie trope but it’s not that far fetched from being doable. Eg from 20 years ago: https://www.theregister.com/2002/05/16/gummi_bears_defeat_fi...
Just want to point out that foreign nationals are apparently sometimes fingerprinted and their DNA swabbed too on entry (according to the travel advice of several governments) - and electronic devices are sometimes searched at the border too. Biometrics are routinely taken for longer stays according to: https://www.themoscowtimes.com/2021/07/01/russia-imposes-bio...
In Russia they don’t use harvested fingerprints to unlock phones. They use wrenches and flat irons to get info and bad windows on higher floors and polonium tea to prevent anyone else getting info.
Here in the US we use both (three types really) for different scenarios; they both have benefits in terms of “quickness” to detect smoke, fire, or heat.
