Fair point on the second one — "AI-generated" is doing more work than it should as a label. It's not autopilot. It reads the resume and builds questions from wh

Fair point on the second one — "AI-generated" is doing more work than it should as a label. It's not autopilot. It reads the resume and builds questions from what's actually there — your stack, your transitions, your gaps. Still a draft tho, you're meant to edit it. On bias — I tie everything to resume content specifically to avoid the name/school inference problem. But have I done a formal bias audit? No. That's a real gap and I'm not going to spin it. Leadership and estimation categories — heard this exact thing from a few teams already and I keep improving this. It's coming. Thank you for the feedback!