Thread

1. hollowonepl · 1790614021 · focus · HN ↗

   AI is not for programmers, they will always complain and given time and resources can of course produce better atomic code. AI is for a combination of a business analyst, architect and product manager in one person who has capacity in coding but decided this is limited role in the whole software engineering lifecycle. With the right attitude and of course certain level of micro management over AI (which they had to do with human programmers as well) they can achieve the same or similar results with much less communication friction, less people to maintain that teamwork and get closer to solving issues they find rather than discussing philosophy of programming and other manifestos of the craft that stopped being prestigious or unique and that’s the key point of complaint when AI is given wrong people to use.

   1. tovej · 1790614162 · focus · HN ↗

      Can they really achieve similar results though? If quality suffers greatly, and velocity eventually slows down due to huge tech debt, isn't that just a worse outcome?

      1. whattheheckheck · 1790614549 · focus · HN ↗

         "Bet on it and start your own company"

         1. hollowonepl · 1790668325 · focus · HN ↗

            If you believe software still can be sold. as I do believe further development of AI may completely redefine buy vs. build capabilities and potential directions.

            My bet is that half of the software developed so far may be obsolete 10 years from now and so the businesses behind them. SaaS specifically.

            1. slohr · 1790734328 · focus · HN ↗

               I'm seeing this right now in my domain. Much stronger lean towards build than years past. Not enough signal yet to really make a call but I suspect a significant downward re-pricing is going to have to happen to convince folks to buy. At the moment the arguments of "your core business isn't software so why build?" or "you'll have to maintain this and there's no one else on the planet that knows it", etc are not working since some of these folks have fully bought in and just say "yeah so, AI will do it." Also the number of features expected, and many non-trivial, is pretty high. wild times!
      2. hollowonepl · 1790668595 · focus · HN ↗

         My 12+ months learning is that quality suffers primarily when I cannot find the right working process to assure it. It's not software development as a process but software engineering as a whole holistic picture across all departments and traditional roles (dev,test,devops/sysops, pm, etc..)

         I also found that what is commonly called "spec-driven development" is just a different CMMI alike model that works better with AI than anything AGILE. which is also nothing new.. just realisation that AI is the ultimate proof agile development/manifestos is a dead end.

         there are many other reflections about the whole process to ensure quality, good architecture, minimizing regressions and providing business continuity and proper, reliable, SLA driven maintenance.

         Actually not that much different than certain human-centric models.. just need to be adapted to the AI reality to truly benefit from it.

         when that is done, from leadership perspective, it is a working process that brings similar outcomes and quality. not exactly cheaper atomically, but faster and brings more scalable opportunity to drive multiple projects within the same resource pool.

         I think that's what is mostly experimented today and that is also of course exhausting traditional developers.

   2. mkleczek · 1790617392 · focus · HN ↗

      That's my recent experience as well. You have to know the domain and have technical knowledge so that you understand (not in every detail) what AI is actually doing and why. You also have to understand what you want and why you want it.

      But once you do, working with AI is like having a very capable team of programmers knowing all programming languages you might need and understanding your goals.

Unofficial Hacker News client; not affiliated with Y Combinator.