This onchain engine runs on memes & RWAs & vibes.

Proof of build by @antoshi & @buttondotfun solves this

IPOs will move on chain.

That’s actually a beautiful idea. Performance and throughput tracking via software metrics is standard practice in tech companies and corporate environments in general. It’s surprising it took this long to see the primitive brought on-chain. I believe @redemptionarcc compared @buttondotfun to MetaDAO for RH, which is a valid comparison to an extent. I wouldn't go quite that far (for now...): while MetaDAO’s extensive due diligence and selective curation set a very high bar, Button adopting a similar framework could position it as the premier launchpad for legit tech and high-conviction projects. And honestly, I’d love to see it. God knows this chain needs fewer larps and rugs. Way too many of them clogging up the ecosystem. The way their Proof-of-Build system works is pretty straightforward: you hook up any token to their setup, allocate supply, then connect your Claude/Codex API. A committed Claude Code hook (two small files in the project's repo) reports the AI token counts from every build session. An oracle proposes a burn of the token each period at a rate of the team's choosing, signed by an AWS KMS key that can only ever send to the dead address. Verified usage is bucketed into hourly periods and combined with shipped output (PRs, releases, deploys) from the team's public GitHub. A burn is then proposed on-chain and serving a 24h challenge window, during which the guardian can dispute it, after which it is executed. Your community can watch you ship in real time. Plus, you can embed a proof-of-build widget on your own site or share the console. Great mechanism for keeping teams accountable and making sure they actually deliver on their roadmap. Naturally, no system is immune to manipulation, and some teams will inevitably attempt to game the metrics with vanity commits or cosmetic prompts. However, the foundational concept of Proof-of-Build is great. As edge cases emerge, the model will undoubtedly be refined, hardened, and iterated upon to minimize exploitability. Instead of evaluating a dev's throughput based on raw lines of code, Button can simply adapt the anti-gaming heuristics and peer-review safeguards that Web2 engineering leads have spent years refining, such as weighing token burns against AST (Abstract Syntax Tree) diff changes, test coverage metrics, code churn/rework metrics (to track how much code is rewritten or deleted shortly after being committed). Or, instead of reinventing the wheel, they could simply use productivity tools like LinearB, Jellyfish, and SonarQube that offer a lot of features. They’ve already lined up partners like @Hedgetradex on RH and @vcdotfun on Solana in the last 48 hours alone. If they keep shipping, locking in strong partnerships, and pulling in real teams, this could do really well. Good luck! I'll be keeping an eye on this one. $BUTTON 0x86e5ac7ec60eeec0a8989f1f1d617ae18c15fa0c