We said MAYBE on August 15, 2026. Not settled — due August 15, 2027.
Read this with the caveat. We tested the engine that produced this verdict against 292 launches whose outcomes we already knew, and could not show it predicted which survived. Some of its data sources were also dead at the time of scoring. The verdict stays up, dated and unedited, because a record you can quietly revise is not a record — but it is worth less than it looked when it was written.
Demand is real and proven by five paying competitors plus growing keywords and active HN interest. Winnability is moderate: genuine pricing and friction wedges exist, but the incumbents are funded and the core AI/device-cloud engineering is heavy, so success depends on ruthlessly narrowing to one cheap-to-build wedge rather than fighting the platforms head-on.
Multiple channels agree: 5 established competitors successfully charging money, 15 HN discussions with high point counts (185/161/132), and growing commercial keywords. Social/Reddit is empty but for a B2B dev-tool that matters less.
Clear wedges exist (developer-first flat pricing, no-auth quick testing, open-source core), but incumbents include well-funded giants (BrowserStack, Sauce Labs, Applitools) and the AI test-generation engineering is genuinely heavy for a solo dev.
The listed wedges (device cloud parity, AI-generated test suites from session recordings, cross-platform mobile) are deceptively capital- and engineering-intensive — running real device farms and reliable test generation is not a weekend build, and well-funded incumbents can copy a pricing wedge overnight. A solo founder may spend months building infrastructure before reaching a payer, which argues for skipping the full-platform ambition in favor of a narrow slice.
What is worth more than this page. The register records what became of 14,962 real launches. No engine has to be right for that to be true.