Back to all discussions

  •   about 1 year ago

Why not use AI to judge projects?

Curious to know since the event as spoken about on CVMTV.

Why not include an AI agent judge?

One that is so robust, the quality delivered could lead to project improvement.

  • 2 comments

  • Manager   •   about 1 year ago

    Hi Rodger,

    At Intellibus we have a Internal Software Engineering Lab that has been experimenting & working on evaluating using AI Agents for Code Reviews & Product Assessment. So far we feel that AI Agent Systems, even with sufficient controls, are not impervious to "prompt hacking".

    We also believe that the Judges at the Hackathon each bring a unique perspective which will be invaluable in assessing hackathon submissions, which AI is not currently able to replicate.

  •   •   about 1 year ago

    Valid point, which proves further how important an AI agent judge, or evaluation for projects that may, or may not involve code, is extremely important.

    In essence, such an AI agent ideally should suited to evaluate code + evaluate a MVP market viability with valuation.

    Currently I have an agent for the latter, but did not include code evaluation, as that would be more detailed after the business evaluation for groundbreaking innovation, viability, and investment score.

    Thank you for the detailed follow up response.

Comments are closed.