This is only going to get worse with LLMs. Now people can "contribute" garbage code at 10x the speed. We're entering the era of the "read only" maintainer focused on self-defense.

I’ve seen it already where someone has set some fully automated agent at the GitHub issues and machine gunned PRs every minute for every reported issue. Likely never looked at or tested by the submitter.

Even if they worked, it would be easier for the maintainer to just do that themselves rather than review and communicate with someone else to resolve issues.

...that assumes LLMs will contribute garbage code in the first place. Will they, though?

The problem isn't that it can't write good code. It's that the guy prompting it often doesn't know enough to tell the difference. Way too many vibe coders these days who can generate a PR in 5 seconds, but can’t explain a single line of it.

That’s 100% the trick to it all. I don’t always write code using LLMs. Sometimes I do. The thing that LLMs have unlocked for me is the motivation to put together really solid design documentation for features before implementing them. I’ve been doing this long enough that I’ve usually got a pretty good idea of how I want it to work and where the gotchas are, and pre-LLMs would “vibe code” in the sense that I would write code based on my own gut feeling of how it should be structured and work. Sometimes with some sketches on paper first.

Now… especially for critical functionality/shared plumbing, I’m going to be writing a Markdown spec for it and I’m going to be getting Claude or Codex to do review it with me before I pass it around to the team. I’m going to miss details that the LLM is going to catch. The LLM is going to miss details that I’m going to catch. Together, after a few iterations, we end up with a rock solid plan, complete with incremental implementation phases that either I or an LLM can execute on in bite-sized chunks and review.

The LLM isn’t contributing garbage, the user is by (likely) not testing/verifying it meets all requirements. I haven’t yet used an LLM which didn’t require some handholding to get to a good code contribution on projects with any complexity.

I heard a quote recently that I really like.

"AI sucks at most peoples' jobs. If you think AI is good at something, chances are you suck at it too."

Tbh, for me there’s basically no difference between a repo with 2k stars and one with 20k.

Stars only matter when there are very few, like if it has almost none, that’s a red flag. Otherwise it’s just noise.

My first project came from a former coworker who moved to a new company. That's pretty much it.

Can't tell you any clever acquisition strategy. For this sort of work you need a critical mass of credibility and connections. The more companies you've worked at, the more people who can vouch for you from the inside. When you're in corpo, you are basically pre-selling your consulting pipeline, before you ever need it.

On a personal note, I quit that hustle, simply because I didn't enjoy having to prove myself every other day to new prospects. Especially since I've been a software engineer for 12 years already. Now just work on my own products that can speak for themselves.

As someone who went to work for a commercial software consultancy company early in my career, I am looking to follow in your footsteps. In consultancy, projects are over for me when my work is done. Rarely do I get to see the results. I want to work at a product company next.

one thing worth pointing out is that the legacy private railways work because they were never nationalized and had decades to quietly buy up land around stations before it was worth anything. That's really hard to replicate from scratch. This model is great in dense cities but even Japan is still struggling with rural lines

Exactly this. And the European case is the opposite starting point. Paris already had 2+ million people when the Metro opened in 1900. You're not building rail to create land value, you're building it because the existing density already demands it. Which is why European systems basically all require public subsidy while the Japanese private lines could turn a profit. The preconditions are just so different that copying either model somewhere else rarely works. IME the people pushing "just do what Tokyo does" tend to skip over this part.

Exactly. Tokyu's model of building a train line from the city center to rural areas and then building suburban developments in the rural areas the line traverses doesn't work in already built-up areas. Hence, there are still publicly-owned lines in areas where that model doesn't work. A great example is the Yokohama Municipal Subway. It is publicly owned and serves areas that were generally already built before the subway line was built.