4 ms·

its completely over for this field. anyone that doesnt think so has their head in the sand, we are 18 months out from the value proposition of a 50-85th percent

its completely over for this field. anyone that doesnt think so has their head in the sand, we are 18 months out from the value proposition of a 50-85th percentile software engineer dropping to the floor

teeray 2y ago
I’ll believe it when I can page an AI at 3AM, go back to sleep, and have it autonomously root-cause and fix a production outage no matter the cause.
- Yenrabbit 2y ago
  What percentage of human software engineers would you guess pass this bar?
  - teeray 2y ago
    Irrelevant. It's greater than the AI's percentage.
  - Capricorn2481 2y ago
    I'd say most of my teammates could do this. It's not a unique skill to debug a prod error. I would say 0% of AI does? What AI can even read production environment vars, find logs in arbitrary parts of the file system, measure usage statistics, and fix the code without causing another bug?
pphysch 2y ago
Code monkeys can essentially be replaced by GenAI (that is, a competent SWE team needs fewer or zero low-skill code monkeys going forward) but the "field" goes way beyond code monkeys. Replacing piles of expensive crappy SaaS with a small team of expert SWE will continue to be a massive arbitrage opportunity for many mid/large orgs.
- StefanBatory 2y ago
  And then the thing is, how do you train new employees? Any juniors will be useless. Year, or two of unpaid internships? It's nothing new. In my country it works like this for doctors or pharmacists - they have to do half a year of unpaid work if they want to get diploma (and I use work and not internship for reason) Because as a junior you don't bring anything of value to company, you're a loss to them.
  - pphysch 2y ago
    I think the field will mature into a traditional specialist field like doctor/pharmacist. Well-paid junior positions may cease to exist, because we just don't need an army of juniors pumping out low-quality code anymore. But experts remain essential. The GenAI economy transition will be very weird. The (US) education system isn't remotely prepared for it. Why pay a bunch of money for a 4-yr CS degree?
MyOutfitIsVague 2y ago
Not without some major advancements in the technology. As good as Claude 3.5 Sonata v2 is, it still can't do more than a quarter of what I do on a day to day basis, and still makes tons of mistakes. I guess if you spin up fresh CRUD apps or greenfield web apps for a living, you might be in danger, because they are really good at doing that. On an established codebase, especially with more than the low tens of thousands of lines of code, their usefulness is mostly relegated to writing boilerplate that can be explained in plain English. I'd be surprised if 85% of programmers were spending their days doing that, though. Even if it was that serious, I don't see software engineers who can leverage LLMs and keep learning with the rest of their field being in much danger. The average software shop that I'm familiar with has a massive backlog of things to be done, and more direction than they have steam. We're pretty far from the point where almost any professional programmer will be saying "We finished all our work and there's nothing left to do".

teeray 2y ago

I’ll believe it when I can page an AI at 3AM, go back to sleep, and have it autonomously root-cause and fix a production outage no matter the cause.

Yenrabbit 2y ago
What percentage of human software engineers would you guess pass this bar?
- teeray 2y ago
  Irrelevant. It's greater than the AI's percentage.
- Capricorn2481 2y ago
  I'd say most of my teammates could do this. It's not a unique skill to debug a prod error. I would say 0% of AI does? What AI can even read production environment vars, find logs in arbitrary parts of the file system, measure usage statistics, and fix the code without causing another bug?

pphysch 2y ago

Code monkeys can essentially be replaced by GenAI (that is, a competent SWE team needs fewer or zero low-skill code monkeys going forward) but the "field" goes way beyond code monkeys. Replacing piles of expensive crappy SaaS with a small team of expert SWE will continue to be a massive arbitrage opportunity for many mid/large orgs.

StefanBatory 2y ago
And then the thing is, how do you train new employees? Any juniors will be useless. Year, or two of unpaid internships? It's nothing new. In my country it works like this for doctors or pharmacists - they have to do half a year of unpaid work if they want to get diploma (and I use work and not internship for reason) Because as a junior you don't bring anything of value to company, you're a loss to them.
- pphysch 2y ago
  I think the field will mature into a traditional specialist field like doctor/pharmacist. Well-paid junior positions may cease to exist, because we just don't need an army of juniors pumping out low-quality code anymore. But experts remain essential. The GenAI economy transition will be very weird. The (US) education system isn't remotely prepared for it. Why pay a bunch of money for a 4-yr CS degree?

MyOutfitIsVague 2y ago

Not without some major advancements in the technology. As good as Claude 3.5 Sonata v2 is, it still can't do more than a quarter of what I do on a day to day basis, and still makes tons of mistakes. I guess if you spin up fresh CRUD apps or greenfield web apps for a living, you might be in danger, because they are really good at doing that. On an established codebase, especially with more than the low tens of thousands of lines of code, their usefulness is mostly relegated to writing boilerplate that can be explained in plain English. I'd be surprised if 85% of programmers were spending their days doing that, though. Even if it was that serious, I don't see software engineers who can leverage LLMs and keep learning with the rest of their field being in much danger. The average software shop that I'm familiar with has a massive backlog of things to be done, and more direction than they have steam. We're pretty far from the point where almost any professional programmer will be saying "We finished all our work and there's nothing left to do".