Most builders give up after their first failed attempt.
@enoch_danijel built 4 different perp DEX models before founding @LemonMarkets - the one that worked.
Here's how he iterated his way to success 🧵
The problem seemed simple: he wanted to long/short meme coins.
But every existing perp DEX model he studied (GMX, Hyperliquid, AAVE-based) was either:
• Too expensive to bootstrap
• Required massive liquidity upfront
• Wasn't scalable for new tokens
So he did what most builders don't:
He rebuilt it from scratch. Multiple times. Each iteration taught him what didn't work. Each failure revealed constraints he hadn't seen before.
His process for each iteration:
1. Research existing models
2. Build for weeks
3. Test with real use cases
4. Discover it doesn't scale
Most people would quit at step 4. He went back to step 1.
His toolkit for rapid iteration:
• AI for breaking down complex white papers and docs
• Twitter search for finding others who've solved similar problems
• Testing assumptions quickly before spending too much time on the idea
The turning point was changing his perspective:
Instead of asking "how do other perp DEXs work?"
He asked "what's the simplest model that solves MY specific problem?"
That led to the P2P matching approach.
His advice for builders stuck on hard problems:
"Don't double down on your failures. Learn from them and keep iterating."
"Do not make the same mistake twice."
"It's not work. It's fun. We're building things people will use."
Innovation often comes from builders who are willing to:
• Throw away weeks of work
• Start over with new assumptions
• Treat each failure as data, not defeat
4 iterations might seem like a lot. But it's what it took to find the right solution.
Watch the full showcase here:
1,844
10
本页面内容由第三方提供。除非另有说明，欧易不是所引用文章的作者，也不对此类材料主张任何版权。该内容仅供参考，并不代表欧易观点，不作为任何形式的认可，也不应被视为投资建议或购买或出售数字资产的招揽。在使用生成式人工智能提供摘要或其他信息的情况下，此类人工智能生成的内容可能不准确或不一致。请阅读链接文章，了解更多详情和信息。欧易不对第三方网站上的内容负责。包含稳定币、NFTs 等在内的数字资产涉及较高程度的风险，其价值可能会产生较大波动。请根据自身财务状况，仔细考虑交易或持有数字资产是否适合您。