"What are the odds Git is the fastest and most efficient way for agents to work?"
Quite high. The odds are quite high.
Also hilarious that the blurb focuses on "saving milliseconds" when that all gets obliterated by the seconds / minutes wasted by llms. Just use git commands man. Or ask your llm to, it's not hard.
I just had a talk (power-rushed a bit) yesterday at Pass The SALT 2026 talking about (my version of) latency hiding filesystems in userspace over the network. My benchmarks, setups, and how to make the magic wawaweewa work.
Basically it went the "sshfs over iroh" approach, and instead of OP's post, it has eager metadata file sharing, and on demand file access. It is still a work on progress, but you can use git juse fine over the fuse mount points without introducing a new tool (edit: but do note that I did not encounter problems with large worktrees, as I do not use them in my flow, and still has some quirks). And still have similar "cold path" access times.
I touch on some insights of hiding the latency, the direct Alice to Bob connection, and has some recorded demos. Goes the generic approach.
You can watch it here if you are interested, but its around 25 minutes.
The first part is about the Post Quantum Crypto, and making direct IPv6 connections work.
The second part is about the filesystems in userspace that hide the latency.
(The second part is around minute 12:00, and I think it is the relevant part to this discussion/ thread).
I don't think we're going to convince anyone to replace git on speed alone. I don't spend that much time waiting for git operations, even on really big repositories.
genuine question: is latency that big of a deal wrt tools for agents? I see this a lot with these AI-native tooling replacement projects, but surely the inference will be orders of magnitudes slower?
I did not measure agents, but lets take the following example:
Assume you have a 10 GB blu ray movie that you want to watch without waiting for upload/ donwload time.
Maybe you host it on a server on another continent, where you got a 200ms RTT.
And your wire at home supports around 500Mbps - approx 62 MB/s.
Thus to Download it fully, would take around: 161 seconds if you use the full wire and on the happy case.
But to actually watch it and skim through it, you do not need to use the full wire, but around ~11 Mbps for a 2-hour 10 GB film, which is like 2% of the wire. The only hard part is random jumps, as each seek is a full RTT.
If you prefetch too aggresivelly you hide this RTT, but if the wire is saturated, and if you have a cache miss, then you will hang and wait until the request goes through. Might be upt to 1-2 seconds. The whole game is to pace yourself such that you stay ahead of the consumption rate, without saturating the wire.
This looks like an interesting design, but I kind of just want to wait 3 years and see if some new VC system emerges victorious rather than try out all these new things
we’re on hacker news! this is the right place for the adventurous people to ‘try out all these new things’ so that the new stuff can emerge victorious :P
From the obviously slop-generated website, to the wild technical inaccuracies like "Git makes you clone the entire history before you can open a single path", to the weird focus on performance when runtime is usually dominated by the LLM, I think I can safely give this a miss. Oh and,
> The commit-message tax.
> Git demands prose on every commit. You burn tokens writing "wip", "fix", and "address review", messages no human will ever read, just to checkpoint your own progress.
Yes, making agents describe their work in a way that is readable to humans and other agents is clearly a bottleneck that must be fixed.
What is this garbage? The only thing I agree with is that worktrees kind of suck (usually once submodules are involved).
Quite high. The odds are quite high. Also hilarious that the blurb focuses on "saving milliseconds" when that all gets obliterated by the seconds / minutes wasted by llms. Just use git commands man. Or ask your llm to, it's not hard.
Basically it went the "sshfs over iroh" approach, and instead of OP's post, it has eager metadata file sharing, and on demand file access. It is still a work on progress, but you can use git juse fine over the fuse mount points without introducing a new tool (edit: but do note that I did not encounter problems with large worktrees, as I do not use them in my flow, and still has some quirks). And still have similar "cold path" access times.
I touch on some insights of hiding the latency, the direct Alice to Bob connection, and has some recorded demos. Goes the generic approach.
You can watch it here if you are interested, but its around 25 minutes.
The first part is about the Post Quantum Crypto, and making direct IPv6 connections work.
The second part is about the filesystems in userspace that hide the latency. (The second part is around minute 12:00, and I think it is the relevant part to this discussion/ thread).
https://passthesalt.ubicast.tv/videos/2026-keibidrop-post-qu...
Assume you have a 10 GB blu ray movie that you want to watch without waiting for upload/ donwload time.
Maybe you host it on a server on another continent, where you got a 200ms RTT.
And your wire at home supports around 500Mbps - approx 62 MB/s.
Thus to Download it fully, would take around: 161 seconds if you use the full wire and on the happy case.
But to actually watch it and skim through it, you do not need to use the full wire, but around ~11 Mbps for a 2-hour 10 GB film, which is like 2% of the wire. The only hard part is random jumps, as each seek is a full RTT.
If you prefetch too aggresivelly you hide this RTT, but if the wire is saturated, and if you have a cache miss, then you will hang and wait until the request goes through. Might be upt to 1-2 seconds. The whole game is to pace yourself such that you stay ahead of the consumption rate, without saturating the wire.
> The commit-message tax.
> Git demands prose on every commit. You burn tokens writing "wip", "fix", and "address review", messages no human will ever read, just to checkpoint your own progress.
Yes, making agents describe their work in a way that is readable to humans and other agents is clearly a bottleneck that must be fixed.
What is this garbage? The only thing I agree with is that worktrees kind of suck (usually once submodules are involved).
[0] https://cursor.com/origin
Is this AI slop or a genuine project?