One thing this thread nails is that bots aren’t a bug, they’re the baseline and that realization completely changed how we approached building a Pumpfun-style token launch portal on Ethereum for a client who originally thought the goal was to beat bots, when in reality the real problem was designing a system that assumes bots exist and still feels fair and usable for humans. Early prototypes looked great on paper, but in testing, snipers, sandwichers, and volume-faking wallets immediately distorted launches, so we stopped focusing on anti-bot gimmicks and instead redesigned the mechanics: commit-reveal style launches instead of open mempool buys, per-block buy caps tied to unique wallets plus delayed liquidity activation and on-chain signals that surface whether early volume is organic or likely automated. The result wasn’t bot-free (nothing ever is), but it materially changed user behavior: humans could actually participate without instantly becoming exit liquidity and serious projects cared more about sustainable traction than fake charts. I think the bigger takeaway, echoing some comments here is that you either build assuming machines will be your main counterparties or you’re building a museum piece; Pumpfun on Solana works because it embraces that reality rather than pretending retail traders are the dominant force. If anyone is exploring a Pumpfun-like model on Ethereum and wants to sanity-check architecture, economics or bot-resilient launch mechanics, I’m happy to guide you.
[link] [comments]
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.