You don't need a software factory. You need an abstract software factory provider which can instantiate concrete software factories without coupling the client to a specific software factory implementation. Have some fucking self-respect.
I have as of now, officially read the entirety of A Song of Ice and Fire.
Surely George will release the Winds of Winter now that I specifically have read the series.
It used to be the case that _eventually_ folks were forced to solve the hard thing or it wouldn’t happen. But now folks can just send Claude off to build a nuclear power plant while they continue chattering about their pet issues.
Jury’s still out, but this seems very bad to me!
Bike-shedding is the idea that people within organizations spend a disproportionate amount of time discussing minor but easy-to-grasp issues like bike shed design while neglecting the much more important, but difficult talks (like designing a nuclear power plant)
You see this all the time in software development. Meetings focused on people’s pet issues like language choice, monorepos, choice of llm etc while avoiding the difficult but crucial issues the meetings are ostensibly about
I've got an agent in a loop optimizing a renderer with the goal to minimize frame times (and tests to measure). It got times down from 88ms to 2ms and allocations down from ~150K to 500. Sounds good, right? Wrong. This is exactly why agent psychosis is a big fucking problem.
As an experiment, I rewrote the Ghostty core render state in Go, with access to identically laid out data structures as Ghostty and the exact same validation tests. I made a purposely naive renderer (simple, correct, but slow). 88ms per frame with 150,000 allocations (horrendous, lol)!
I then kickstarted a Ralph loop to bring the frame times down. I told it it can't modify input data structures or the public API or tests (they're correct), but it can do anything else it wants. It got to work.
It has worked for about 4 hours. I've spent around $350 on this experiment so far. The results?
88ms => 1.5ms
150K allocs => ~500 allocs
Incredible right? Nope.
My hand-written renderer I ported has frame times (same benchmark) of ~20us (0.020ms) and 0 allocations in the update path.
This is the problem with psychosis and lacking systems understanding. If you don't understand the system, you're going to accept that this is an incredible result. If you understand the system, you'll see better solutions immediately and can do roughly 75x better on throughput.
The people who blindly trust agent output are in the former camp. They're sheeple, overdrinking from a fountain of mediocrity.
Standard disclaimer: I use AI all the time. I like AI. The point I'm making is to not blindly accept results. Think. Analyze. Learn.
All the vibecoding startups: AI will write all the code and put most devs out of a job.
AI *infra* startups: record growth, our customers are vibecoding/AI startups who use our compute platform, databases etc. 😅
Even the vibecoding companies aren’t vibecoding their own infra.
my results on AI for autonomous progress:
we ran codex+claude for 43 days straight to build a System Verilog compiler/simulator
https://t.co/f0tT0cBtOM
Struggling with minibatch noise in Stochastic Gradient Bayesian Inference? Want your chains to naturally run on stochastic hardware?
Introducing — Stochastic Gradient Lattice Random Walk!
New work from @NormalComputing in collab with @zierhjmensch, @adnarim066, @wellingmax, and the dream team at normal computing, @Sam_Duffield, @MaxAifer and @ColesThermoAI.
Wasted an hour this week because the custom firmware Claude built for my split keyboard to support Dvorak/homerow mods somehow also disabled Bluetooth
Like and subscribe for more productivity tips
Whenever my pill organizer is empty I refill all 7 days and then pull out an extra set for the current day. This means that every 8th week I don't have to refill it. Please like and subscribe for more productivity hacks.