How does everyone setup their local computers for dev work?
Does anyone find their machine die when you run everything local.
Ai code gen, docker, api servers etc? I have 2 instances of windsurf and swap between these to then run the mode api servers.
Currently computer is a a mac mini (m2) with 8 gig and a latest mac air. (Yup old)
.. the mac mini is sooo slow and needs an upgrade.
What macs has everyone found found to be good
I was looking at the latest mac mini
10-core CPU
10-Core GPU
24GB Unified Memory
512GB SSD Storage
IS the next mac mini , Worth the extra price?
12-Core CPU
16-Core GPU
24GB Unified Memory
512GB SSD
Or should I go mac studio, is it really worth the extra $$?
Trending on Indie Hackers
Ok, I couldn't wait for the m5 mac mini... So went with the m4 mac mini with 24 gig ram.
the difference is HUGE, soo much faster...
I don't believe I am going to say this, but maybe too fast? :)
I got so used to working on my mac mini and my laptop, that I would work on 2 projects at once... While I was waiting for one project to do its thing, I would switch the other computer to work through something there....
Now the new computer works so fast, there is nothing to wait for...
Thanks everyone, solid advice. Going to wait to see when the new mac mini m5 is released and get that with alot of ram.
We learned the hard way that tooling matters less than consistency. Once everyone had roughly the same base setup, a lot of “works on my machine” issues just disappeared. The biggest improvement for us was treating local dev as something reproducible, not personal
8GB is definitely the bottleneck here, especially with Docker and AI tools like Windsurf which are quite memory-intensive. We've found that for a modern stack (Python/Flask, Docker, AI agents), 32GB is the sweet spot to avoid swap lag. If you're upgrading the Mac Mini, definitely prioritize RAM over core count if you're hitting that 'system out of memory' wall frequently.
Totally feel this — running Docker + local servers + AI tooling on 8GB starts to choke fast.
One pattern I’ve seen in dev environments is that memory headroom matters way more than CPU clocks for local workflows. Docker swarms, language servers, runtimes, and local databases all add up quickly, and once you hit swap your whole machine feels sluggish.
On Macs specifically, the 24GB unified memory tends to make these workloads feel predictably fast across a mix of containers and services without hitting swap, whereas the base 8GB setups often hit bottlenecks early.
If you find yourself constantly juggling services and restarting because of memory pressure, the Mini with 24GB usually gives the best bang-for-buck before considering something like a Studio, which is great but often overkill unless you’re also doing GPU-heavy work or large model training.
8GB is definitely the bottleneck here, especially with Docker and AI tools running simultaneously. Docker eats RAM for breakfast
I think the leap to the 24GB Mac Mini will feel massive compared to what you have now. Unless you're doing heavy video rendering or 3D work, the Mac Studio might be overkill for web dev. I'd go with the 10-core/24GB Mini and save the extra cash
Mac Mini's good enough. I have a Mac Mini with M4 Pro chip (Though not for dev work but I can see it being useful for docker work + API servers).
Atleast a base Mac Mini M4 is actually good enough for your work. I wouldn't go studio unless you have something REALLY heavy going on.
At that point I’d honestly look at a Mac Studio, even used, imo memory headroom matters way more than chasing the newest Mini. Those workloads stack fast and the Studio just feels less fragile under constant load.