I'm buying into this workflow more the more I use it, but the real gamechanger is (a) parallel threads in worktrees, with (b) enough lifecycle hooks to treat them similarly to spinning up a VM.
Specifically for me that means that after I create a worktree I get some local config files copied over and Postgres duplicating my local dev and test databases so I can test in isolation, and then when I close out a worktree it deletes those databases.
The best at that that I've found is Conductor, but I can't use it at work because we only have Copilot and they're locked to a Claude/Codex backend. Arbor is close, but it's not under as active development and has a lot of rough edges. Opencode GUI has create hooks but not teardown.
If Zed can hook that up _and_ also keep its great editor roots, that'll definitely be a game changer.
This is helpful to know - we're working on adding more agents, Copilot and OpenCode harnesses are among the most popular requests.
We also recently built an escape hatch. If you turn on Settings → Experimental → Big Terminal Mode you can create new terminals in the center panel (with ⌘⇧T) and use any agent you'd like (Copilot, OpenCode, etc). It isn't the best experience because you don't get notifications etc (yet), but at least it lets you use the harness you'd like until we build out the first-class UI for it.
Send me feedback anytime, I'm charlie@conductor.build.
I’m interested in these types of tools, but the biggest issue I am currently having in my workflow is manually verifying everything is working. Tests and stuff are nice, but typically if there is a bug, the AI agents enshrine it in th tests.
However with worktrees I am not really able to easily copy secrets, etc to run my app, ports conflict, I end up with a bunch of separate dbs and services, etc.
Does conductor help with this? Have you all found any useful ways of making this easier or more automated?
I was an early user of conductor and used it a lot (like from maybe oct to Jan). But then there was some bug where maybe it wouldn’t release file descriptors or something where my laptop needed to be rebooted twice a day. So I stopped using it months ago.
But I’ve tried to reinstall it since and it just gets stuck in a weird infinite loop.
I liked conductor though. Hope you are able to fix those bugs and I can try again in a few weeks.
Howdy, love your product. My coworker and I have literally been begging our coworkers to give it a try. To me, it was the key tool to 'get it' (coming from Cursor) when going from single threaded agentic dev to parallel agentic dev. Please keep up the great work.
Conductor is really nice!
A feature I would really like would be to be able to automatically configure “direnv allow” in the different shells as I use nix & direnv all the time
Unless I misunderstand something but why not just roll a couple of helper shell scripts that create a new git worktree for you, copy over a local .env file (or any other config file) and populate it with ports/variables that are unique across all worktrees (to avoid collisions if you're on localhost, but you could also do this with Docker) and a helper script for tearing down a worktree after you've merged the changes back into main. I'm even supplying unique chrome debug ports and unique temp user data dirs for isolated automated testing in chrome via chrome devtools mcp. Like, I'm not sure why an external library/tool is needed for this.
Oh, yeah 100% I'm putting up scripts for them, the conductor hook is "run the one script" for setup/teardown.
The difference is removing friction, having a UI that shows me what's set up, that I don't need to hit the filesystem or git status to check in, that gives me _direct_ access to a shell/etc in that worktree path.
In VSCode, I use https://github.com/jackiotyu/git-worktree-manager for the same purpose - the extension has before create/before destroy WT hooks which you can run anything from. Mine symlinks workspace file from main checkout, installs packages and copies over some files. Very handy.
You should check out Ouijit [1] - I use it regularly for work and it's nice because it focuses on the environment that you want, and just gives you a shell that you can use any tooling in, as well as VM isolation per worktree if needed.
No need for any AI-specific tool, this is exactly what devcontainer is for! Just tell your agent to use devcontainer up (and docker compose down the other way).
Devcontainers always disappointed me. The sales pitch is that everyone uses the same container, but that's not accurate. Everyone builds a container from the same config and it'll be similar, but it takes a ton of effort to make sure it's identical.
The idea that a devcontainer gets built on-demand instead of checked out like 'docker pull ..." has always felt weird to me. It's so close to being awesome, but ends up being barely useful.
Or maybe I'm wrong. Is there a way to checkout an immutable devcontainer?
tried parallel agents for a sprint and bounced off it. the worktree dance is fine, real blocker for us was test data isolation. scoped postgres schemas per branch worked, but reasoning about which agent broke teh shared migration when three of them touch it got old fast. we just run one agent at a time now and go for a walk.
I've been extremely happy with Arq https://www.arqbackup.com/ for several years as a quiet backup solution, bring your own storage. I've done a few small restores and it's been just fine, and it automatically thins your backups to constrain storage costs.
Managing exclusions is something to keep vaguely on top of (I've accidentally had a few VM disk images get backed up when I don't need/want them) but the default exclusions are all very reasonable.
Second vote for Arq, still on the trial but I have zero issues.
I'm using B2 as the backend, ironically, along with a Hetzner Storage Box. It just runs in the background, has decent defaults for "Don't backup useless crap" etc.
I'm still debating whether to get the single purchase version or pay $60 a year for 5 computers + 1TB of cloud storage.
Both players choose a card. Players then in turns reveal their card, and if Check, make another choice. The player first revealing Checkmate wins if their opponent's currently-chosen card is also a Checkmate.
But then this just gives the win to the first person to open their card, since in that round they had both selected Checkmate. Or, you have an incentive to rush to open your card when you know you've selected Checkmate, as you want to be the first one to open.
Maybe I should've worded differently for clarity, the game doesn't go forever:
The player first revealing Checkmate ends the game. They win if their opponent's currently-chosen card is also a Checkmate, otherwise the opponent wins.
In the proposed game above, there is no rounds, just alternating plays, in which you have to select you play before the other player announces their play, then swap and repeat
So both players select their cards, then player 1 announces, then player 2, then select, then player 2 announces, then player 1? This seems a bit limiting, as you can't really select Checkmate on the play where you don't reveal first, because you only stand to lose.
Yeah, but what stops P1 from DDos'ing and picking checkmate each time?
If P2 picks check the first time, then they're done. At any point after if they pick checkmate, since P1 has checkmate selected they will reveal it and P2 will lose.
You're assume if someone picks 'checkmate' and the next player picks 'check' the games is over and the checkmate selector loses. I assumed that it means you treat it like 'check' 'check' and continue playing. But neither is actually specified in OPs post.
But let's assume it's your rules. Then winning is easy, just never pick checkmate. Literally never. As soon as your opponent picks it, they lose.
So is war (the card game), but people still play it
I think the proposed game has that both of you lose, like tic tac toe. The only way to win is to checkmate as described. Although it is a memoryless game as proposed, so all options (restart, continue, end) are indistinguishable. Maybe if you win, you go again?
Anyways, the game seems to be described to be the equivalent to the political doctrine of mutually assured destruction. Also a terribly designed game.
> with most EVs, you can drive 500 miles with one 30 minute charge
I'm just gonna call bullshit on that one. https://ev-database.org/cheatsheet/range-electric-car looks like it's got a bunch of recent vehicles, and pegs average range at 390km or 242mi. The _highest_ range there is 720km or 450mi, and a 300KW charge runs it 10-80% (+500km range, not miles) in a half hour.
If you _don't_ have the absolute best range + infrastructure to support the charge at that rate: I've got a 2020 Kia Soul w/ 64kWh battery and lines up with the 390km range rating. Did a road trip last year. My charger caps out at 73kW or so on a DC charge, and a charge at that rate (40%-80%) gave me ~150km in a half hour. 10-80% is ~220-250km and takes an hour.
Now, I did the road trip with two kids and a dog, so an hour's potty/walk break every 2-2.5h driving worked out for us, but I don't think that's entirely generalizable. I do also agree that unless we're road-tripping, it's a nonissue. We put a level 2 charger in the garage, and plug in overnight once or twice a week and no range stress at all.
I think they assumed a Tesla Model 3 or Hyundai Ioniq that has 300 miles of range. You leave home with 80% or 100% charge. Stop at a supercharger for a half hour when you have 50 miles left, which will get you back up to 80% battery.
I see this as not just internal API/architecture/code documentation, but product documentation too. We maintain internal docs about how our product is used for our support, implementation, and sales teams to reference.
Right now it's hosted externally (in our "blessed" knowledge base) but if it could be pulled into the repo, and we set an AI reviewer on every pull request to sanity check that if the changes we're doing have a material impact on the feature as described in those docs that it should be flagged (or changes proposed) that'd be a nice win for keeping them up-to-date, and it's easy enough to publish markdown as html or even script an update to the canonical site when we merge to main.
The US decided (and Canada followed) that daylight time was more correct for the larger portion of the year, presumably it's easier to transition the remaining 4mo to daylight than it is to move 8mo to standard.
But also, all the opinion polling (business and individual) was like over 90% in favour of year-round daylight time, so here we are.
> The US decided (and Canada followed) that daylight time was more correct for the larger portion of the year, presumably it's easier to transition the remaining 4mo to daylight than it is to move 8mo to standard.
How is transitioning permanently to one easier than transitioning permanently to the other?
How to transition to permanent DST: wait until we are in DST and then stop switching.
How to transition to permanent Standard time: wait until we are in standard time and then stop switching.
If you adopt permanent DST, the there's a 1 hour difference between the current clock and the future clock for 4 months, and nothing for 8 months. If you adopt permanent ST, the difference between the current clock and future clock is 1 hour for 8 months and nothing for 4 months.
It's a 4 month-hour difference over the year, instead of an 8 month-hour difference.
Personally, I'd prefer standard time, but having all days be 86400 seconds is a pretty great improvement over status quo. I find what most people really would like to change is the amount of time with sunlight in the winter, especially the more north they live... but changing the clock doesn't change the number of hours of sunlight; Vancouver, BC just doesn't have much sunlight in the winter.
If we assume that the ideal time for 8 months of the year is DST and for 4 months is standard, but we want to eliminate the switch, then permanent DST gives you only 4 months out of the ideal timezone rather than 8.
The difference is how they're consumed you don't sit down on Netflix and say "put some scifi on shuffle for 8h", you sit down and choose a show.
If you're the kind of person who would manually queue up 100% of your songs for the day then Spotify Generic songs aren't an issue. If you just hit a "2020s R&B" playlist and go that's where it feels more sketchy.
Are you looking for Crossover? It's a bit annoying to not run Steam natively (no cmd+H to hide, etc) but it's got a lot of support. Performance is decent on my M2 mini, and even cross-platform stuff like Baldurs Gate 3 is comparable performance to native.
Especially anything that Mac Steam natively calls out lack of 32bit support has good support.
Sadly, that's not true—for instance, I was trying to run the Shadowrun Returns series the other day, and while it launches, it will hang indefinitely when you try to actually start a game. (M4 Max)
I previously played through Returns, Dragonfall, and part of Hong Kong on Mac before the 32bit-apocalypse.
Nah, nothing like that. We explored shipping Proton for macOS early on, but decided it wasn't where we wanted to spend our time, so we removed it[1] to focus on Linux. There's only so many hours in the day, and supporting two platforms is a lot more work than one.
Some crossover games perform better than the native ports. I play Path of Exile on Mac using the Windows client with a translation layer, and it plays better than the native release.
Specifically for me that means that after I create a worktree I get some local config files copied over and Postgres duplicating my local dev and test databases so I can test in isolation, and then when I close out a worktree it deletes those databases.
The best at that that I've found is Conductor, but I can't use it at work because we only have Copilot and they're locked to a Claude/Codex backend. Arbor is close, but it's not under as active development and has a lot of rough edges. Opencode GUI has create hooks but not teardown.
If Zed can hook that up _and_ also keep its great editor roots, that'll definitely be a game changer.
reply