Rajko Zagorac 🛟

EVM Dev & Security Researcher | Eng @DeFiSaver | Views = mine

[Rajko Zagorac 🛟 @rajko_z]

@apoorveth Meanwhile, Proton users seem safer

No one can get the link since it's down:
https://t.co/dG7jcldY3D

[Rajko Zagorac 🛟 @rajko_z]

@letsgetonchain Definitely one of the most innovative designs out there, although it does come with more complexity and risk per code written, imo

[Rajko Zagorac 🛟 @rajko_z]

I’ve been studying different DeFi protocol designs lately, so I decided to fully break down how @0xfluid works from a technical point of view.

40+ pages on liquidity layer, vaults, ticks, branches, and liquidations:
https://t.co/1pY9oTacAv

[Rajko Zagorac 🛟 @rajko_z]

@_nikolajankovic @0xfluid Doing my part to fix that 🫡

[Rajko Zagorac 🛟 @rajko_z]

@SHIknowz @0xfluid Glad you like it. Definitely expect more in the future 🫡

[Rajko Zagorac 🛟 @rajko_z]

@defi_alex23 🫡

[Rajko Zagorac 🛟 @rajko_z]

@NPalinkasevic @defi_milos Let @defi_milos cook with the whiteboard, I prefer the dark room.
Good gym material between sets though 🫡

[Rajko Zagorac 🛟 @rajko_z]

@_nikolajankovic Too heavy for a Monday morning, but definitely weekend material.

Tnx for sharing, ser 🫡

[Rajko Zagorac 🛟 @rajko_z]

@litocoen @0xfluid @letsgetonchain So we all got nerd sniped by Fluid. Love it

[Rajko Zagorac 🛟 @rajko_z]

@defi_milos Not a classy whiteboard like yours, ser, but it'll do 🫡

[Rajko Zagorac 🛟 @rajko_z]

@Mardeni01 @0xfluid 🫡

[Rajko Zagorac 🛟 @rajko_z]

@letsgetonchain @0xfluid 🫡

[Rajko Zagorac 🛟 @rajko_z]

@smykjain @0xfluid Must be so satisfying seeing those whiteboard sketches materialized on-chain. Tnx ser, and tnx for building Fluid 🫡. We really need more innovation out there

[Rajko Zagorac 🛟 @rajko_z]

4/ Inspired by @RareSkills_io

If you love RareSkills deep dives, you might love this one as well.

[Rajko Zagorac 🛟 @rajko_z]

3/ Why write this manually in 2026?

I did use LLMs at some points to understand the code better, but everything in these 40+ pages was written by me.

Still think that writing manually and struggling through code is the best way to build intuition.

[Rajko Zagorac 🛟 @rajko_z]

@pashov Yes, audited code, but hopefully no highs or criticals. Few mediums we were already aware of, and a number of useful issues that won’t result in direct code changes but will help us improve the doc and clarify assumptions

[LonelySloth @lonelysloth_sec]

You can’t have a lasting competitive advantage if LLMs are in your critical cognitive path.

Whatever you do will be at best limited to the same capabilities and constraints as what everyone else is doing.

You’ll find the same bugs as everyone else, build the nth clone of the same app, trade using the same strategy.

Using an llm while keeping it out of the critical path isn’t easy.

If your bug leads are LLM generated — you will be constrained by the bugs that can be found by LLMs. No amount of prompting or composition will give you more than a small margin over your competitors that tends to collapse to zero.

If all your code is LLM generated you are constraining your product to the universe of code LLMs can write — which is the same code your competitors can easily write.

Even adding human input and review doesn’t necessarily change the picture. No amount of context, human generated hints, or iteration, can force a model to produce something that diverges too much from it’s training data.

The cost of doing something through an LLM scales exponentially as you move further away from sample. That’s not something the next model will fix. **It’s a mathematical necessity**

Past a certain level of novelty making the model do the work requires more human mental work than just doing it yourself.

But you take a long time to notice it, because the LLMs nudges you back to doing what it’s good at. For coding it just keeps writing something that isn’t exactly what you asked and you keep telling yourself it’s close enough.

For finding bugs the cost shows up as not finding the bugs as the model keeps looking for typical vulnerabilities. For s bug hunter it means spending compute without revenue. For a protocol it can mean getting rekt.

You have a brain evolved over millions of years with priors that match the real world in a way a statistical model of text cant.

Your brain can legitimately extrapolate beyond what others thought in the past in a way that is impossible **even for an LLM with arbitrarily large compute power**

Use it.