Pillar 04 · original research

The Lab

We build real apps with AI, instrument every step, and publish the raw numbers. Experiments, teardowns, and the honest data behind the hype — methodology included so you can reproduce any of it.

42

apps built

120h

logged on camera

14

tools tested

Start here A 3-read path into how we run the experiments

01 →

How we test: the build rubric

The spec, the scoring, the stopwatch.

02 →

Bolt vs Lovable vs Y Build

The flagship head-to-head, fully timed.

03 →

What AI can & can't build in 2026

Patterns from 42 real builds.

All experiments

03

Flagship · Compare × Lab

We built the same app in Bolt, Lovable & Y Build — here's the data

Jun 18, 2026 · Marcus Tan · 11 min

05

Lab note

What you can (and can't) build with AI in 2026

Jun 1, 2026 · Dana Levin · 9 min

Read →

04

Lab note

We let 5 AI agents refactor the same codebase

May 24, 2026 · Marcus Tan · 13 min

Read →

02

Lab note

The real cost of a "free" AI build, 30 days in

May 16, 2026 · Dana Levin · 7 min

Read →

Every experiment runs on Y Build

Run your own build — free to start

Keep exploring