• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI
    AI

    Shital Shah Sees Prompt Gap in App Building Contest

    Research engineer notes CS backgrounds produced stronger AI prompts.

    SS
    1 Source, 26d ago, first seen 26d ago

    TLDR

    Shital Shah, Principal Research Engineer at Microsoft Research AI Frontiers Lab, described an internal app building competition. Participants with computer science or programming backgrounds created markedly better prompts than those without such experience. Shah observed that the developers anticipated where models would fail and adjusted prompts in advance. The source line attributes this observation directly to Shah and does not report outcomes, winner details, or further results from the contest.

    Combined views

    936

    1 Source, first seen 26d ago

    Combined views

    936

    1 Source, first seen 26d ago

    4 likes
    4 likes
    1 saves

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    1 saves

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet

    1 Source

    @sytelusWe had a little app building competition between people with cs/programming background and the ones without. When I looked at prompts on both sides, the difference was night and day. The people with cs/dev background were able to predict where model might not perform and preempt that. You would think that this would only cause more iterations for people without cs/dev background. That unfortunately isn’t true. One key observation is that people with hands on coding experience were able to refine prompts constantly so that code stays manageable between iterations. In effect, their iterations were faster and they could do more of them before hitting diminishing returns. This is very small sample so don’t want to generalize too much but if you are studying cs, engineering, physics, math you should continue doing that without any fear of AI. The concepts and skills you learn will give you a ton of advantage when you switch to “manager” position and just instructing AI to do things.

    1 Source

    @sytelusWe had a little app building competition between people with cs/programming background and the ones without. When I looked at prompts on both sides, the difference was night and day. The people with cs/dev background were able to predict where model might not perform and preempt that. You would think that this would only cause more iterations for people without cs/dev background. That unfortunately isn’t true. One key observation is that people with hands on coding experience were able to refine prompts constantly so that code stays manageable between iterations. In effect, their iterations were faster and they could do more of them before hitting diminishing returns. This is very small sample so don’t want to generalize too much but if you are studying cs, engineering, physics, math you should continue doing that without any fear of AI. The concepts and skills you learn will give you a ton of advantage when you switch to “manager” position and just instructing AI to do things.