> The real secret to agent productivity is letting go of your understanding of the code and trusting the AI to generate the proper thing

The few times I've done that, the agent eventually faced a problem/bug it couldn't solve and I had to go and read the entire codebase myself.

Then, found several subtle bugs (like writing private keys to disk even when that was an explicit instruction not to). Eventually ended up refactoring most of it.

It does have value on coming up with boilerplate code that I then tweak.

You made the mistake of looking at the code, though. If you didn't look at the code, you wouldn't have known those bugs existed.

fixing code now is orders of magnitude cheaper than fixing it in month or two when it hits production.

which might be fine if you're doing proof of concept or low risk code, but it can also bite you hard when there is a bug actively bleeding money and not a single person or AI agent in the house that knows how anything work

"Magic Leap raises $827M in Series C". Aged like wine.

They're still doing well today, right? IIRC they provide the lenses for Metas in-development AR glasses

This is great! Coincidentally, I just started replacing my collection of bespoke security bash scripts with an app like yours. WIP here: https://github.com/leolimasa/age-vault

We all keep reinventing the same thing :)

Yeah, I run macos for the same reason.

However, I went back to linux on my personal laptop (nixos on my case) and I am pleasantly surprised how many things now just work.

The only thing that still annoys me is the laptop not sleeping properly and therefore using too much battery power when idle.

It has made great strides on the last two or so years.

This echoes my sentiment that LLMs are higher level programming languages. And, as every layer of abstraction, they add assumptions that may or may not fit the use case. The same way we optimize SQL queries by knowing how the database makes a query plan, we need to optimize LLM outputs, specially when the assumptions given are not ideal.

ORMs are one of those things that make sense for really tiny projects but fail to scale once complexity settles in.

Exact opposite experience.

“SQL strings are one of those things that make sense for really tiny projects but fail to scale once complexity settles in“

Large projects require reuse, composability and easy refactoring. All things ORMs excel at.

On a small code base it is easy to rename a column or add a column, etc.

On a large code base with already 100s of queries using that table, without an ORM it isn’t as straightforward to update all references and ensure that ever place consuming that table has the new column info.

Type checking is indeed an advantage of ORMs. You pay for it with object relational impedance mismatch. That impedance grows as your schema grows.

In my experience, the way to get the best of both worlds is to use a query builder as opposed to a full ORM.

I love doing research. I published a minor unimportant paper in undergrad and had a blast doing it.

Then at graduation I was offered a well paid job in the industry. Decided to pursue it as opposed to spending 5-6 more years in academia looking for grants.

Would love to go back and get a PhD, but the economics just don't make sense for me. For now, it's a retirement plan.

>What does windows 7 do that windows 10 doesn't? Why does the same web page need 60MBs to load when it only need 1-3MB 10 years ago.

Ads. And tracking code to serve you ads. And AI - that collects your prompts to serve you more ads.

\s

This is great! How does it compare to silverbullet (https://silverbullet.md/)?

As a manager, if upstream changes priorities on me but what we’re working on is almost done, I just go ahead and finish it anyways.

When they eventually switch back to the original thing they are always surprised to know it’s been completed.

Bad tactics. You let them make stupid mistakes with impunity. That way they'll never learn.