I built Proofd (https://proofd.ai) after researching AI displacement data and realizing that job-title-level predictions are nearly useless. Two Marketing Managers can score 78 and 31 depending on their daily task mix. Well, to be really honest -- I was trying to understand the path for myself first :)

Proofd scores your actual tasks across 5 dimensions (repeatability, data dependency, AI capability, human interaction depth, accountability tolerance) using O*NET occupational data and METR autonomous capability benchmarks.

Built with AI obviously. Looking for feedback on the scoring methodology.

Free instant score. Full task-by-task report for $29 (launch price).

For HN commenters -- giving out reports for free to those who can honestly give feedback. Just put your HN username somewhere when submitting the form, don't make the payment, and just comment here. I'll see it and will send over a free report for you to check out!

Curious what HN thinks.