Ask HN: What's your favorite "what would SWEs do in 1-3 year from now?"
LLM driven stacks by Anthropic and OpenAI are aiming for a monoploy on labor replacement by driving Claude Code, Codex development at rates never seen before, there would be likely a reordering of what SWEs do in the near future(1-3 years).
What's your futuristic version of how this would turn out? Try justifying your answer e.g. by citing previous re-organization of labor during such upheavals, applying economic/market theory or precedent.
My favorite one(right now) is: As traditional SWEs jobs lose pricing power, compensations for SWEs in traditionally non-tech companies(think coca-cola or nike) will become more stark, like Sales commissions based on being able to communicate effectively with AI to steer it or solve bugs created by AI will be bid in a highly competitive market. The best solution will be tested and the bounty winner will go home happy. STEM based immigration to US/UK will fall off a cliff. The most successful of IT companies will go through a series of M&A and competition will intensify in a second race to the bottom on cost and margins will compress.
In larger tech companies, AI will need a lot more steering and here things would be less stark, engineers will make new features requiring to steer AI more often.
Market wise, Private Equity(Direct lending/BDCs) will see a renewed focus on acquiring companies to use AI to make things ever more efficient —the large number of growing private companies will be lapped up by them — and investment in this sector will increase evermore.
Please add your own fav. scenario :) I'm nowhere to be in the big league, but with my "play" with AIs since it's beginning I don't think AIs are still there to replace en masse like you think it will, gradually but I believe eventually it will replace. Also my another opinion is that large amount of One man companies will start to raise as the people who knows how to "play" with AIs will eventually achieve goals they couldn't before due to barriers.