Founder @ Vibe Coach™
8.4K posts
Founder @ Vibe Coach™
@TheChefnshape
Founder of Vibe Coach — a platform connecting businesses with skilled vibe coders to solve real problems using modern AI-driven development tools.
all I know is, when @MemeLiquidio showed up all of sudden there was a ton of development happening from other teams after they launched mainnet lets face it, everyone has devnets and never have had a real working product. ML did, but there was bugs in the original percolator code, not their fault. They fixed a lot of it and obviously made Toly aware because he has been updating the percolator prog since ML created a v2 placekeeper. I think a lot of you are uninformed and this constant competition is immature at this point lol only time will tell who the real builders are and who actually comes with a working mainnet you have to ask yourself, why did Ml start to build in private repositories my guess is that other teams were using their open source code and passing it off as their own proof never lies and all I see now is a bunch of teams promising the world and under-delivering for months now a lot of teams are just in a constant ai slop circle that create solutions that then create more problems, the root problem or deliverable never coming to fruition its sickening at this point to watch you all fight over who is right and who is wrong lol i say stfu already and let the best percolator fork win
If you are not pushing as much as me into GitHub, ngmi
Don't trust. Verify. If someone is building a permissionless perp DEX and asking for your attention or your capital, you should not have to take their word for it. Here is what you can actually check yourself. First, is the code open source? If there is no public repo, the conversation ends there. No code means no verification. Walk away. If the code is public, clone it. Point an AI agent at it. Ask it to read the risk engine, the liquidation logic, the fee split, the oracle integration. A capable coding agent can tell you in minutes whether the code does what the team claims it does. Run the tests. A serious protocol has a test suite. If the tests do not pass, that tells you something. If there are no tests, that tells you more. Then go further. Run Kani proofs. Kani is a formal verification tool for Rust. Instead of just checking that your specific test inputs pass, it checks that mathematical invariants hold across every possible input the program could ever receive. Things like: can a liquidation ever leave the engine with more debt than it started with? Can a fee split ever pay out more than was collected? Can a funding settlement violate the zero-sum property? These are not questions you can answer with unit tests alone. A unit test checks the cases you thought of. Kani checks all of them. If a risk engine has Kani proofs and they pass, that is a fundamentally stronger guarantee than a test suite. If a team claims their protocol is safe but has no formal proofs, their security is based on hope. Look for an audit too. A real third party audit with findings published. Check what was found. Check whether it was fixed before launch. And do not interact with anything on mainnet that has not been audited. Devnet exists for a reason. If a protocol is pushing you to use mainnet before a public audit is out, that is your answer. The tools exist to verify this stuff yourself. Use them.
💹💹💹