01 — Position

I read protocol code for a living — and I can prove what I found, in runnable tests.

2,837Lines of production Solidity authored
9DeFi protocols integrated at contract level
16Protocols put through structured security review
12Audit methodology modules written

Counts come from my own repositories and can be verified directly on request.

02 — Evidence

Selected work

Protocol security review

Foundry PoCs · fork tests
16 protocols tracked to a verdict

Structured review across sixteen protocols — among them Hydration, Alchemix v3, Origin, Maple, Fluid and Hashflow — each one tracked to a written verdict rather than left open. Where the review produced a critical-class finding, it produced a runnable Foundry proof of concept alongside it: a test that reproduces the defect against forked state, not a paragraph asserting one exists.

The discipline that matters here is the one that closes files. Most of those sixteen ended in no submission-ready finding, documented with the areas checked and the reasoning that dismissed each, so the work is auditable and nobody repeats it by accident six months later.

PoC reproduces Fork testing Invariant review Written verdicts

Flash-loan execution contracts

Solidity · Foundry
arbitrage · liquidation · refinance

Executors that borrow, route and repay inside a single transaction across lending and AMM venues — arbitrage, liquidation and collateral refinancing — with profit floors enforced on-chain, so an unprofitable attempt reverts rather than settling at a loss.

Integrated at contract level against Morpho, Aave, Compound/Comet and Euler on the lending side, and Uniswap, Curve, Aerodrome and Slipstream for routing, with Chainlink price feeds.

MorphoAaveComet EulerUniswapCurve AerodromeChainlink

MPC dark pool

Python 3.11+
32 tests · zero runtime deps

A midpoint dark pool for tokenised equities where resting orders exist only as secret shares split across independent nodes. Matching opens one bit per attempt; execution happens at the oracle midpoint, so limit prices are never revealed to anyone — the venue included.

Verified against a plaintext reference implementation with 400 randomised MPC-versus-plaintext equivalence checks, holding the cryptography to the same result as the naive version on every input.

Secure multi-party computation Additive secret sharing Order matching Working MVP

Audit methodology toolkit

12 modules · documented, licensed

A written system rather than a habit: bug-class taxonomies, a static-analysis grep arsenal, Foundry PoC construction patterns, triage and report standards, chain-specific hunting guides, and an MCP integration for Solidity review.

It exists because reproducibility is the job. Anyone on a team can follow it and reach the same verdict on the same code.

Bug classesStatic analysis PoC constructionTriage & reporting
03 — Method

Measuring what others assume

Finding the hole in my own instrument

A scanner I built reported that competitors were bidding roughly 0.000 gwei for the opportunities it tracked — which would mean nobody was competing. I treated that as suspect rather than as good news, because searchers commonly pay block builders through a direct value transfer the standard metric cannot see.

Confirming it normally needs a trace-enabled node, which I did not have. So I measured the same money a different way: the fee recipient's balance delta across the block, minus the priority fees actually charged, corrected for ordinary transfers. Whatever remains arrived through a channel the usual measurement is blind to.

Measure, 39 consecutive mainnet blocksResult
Blocks carrying a hidden builder payment14 / 39
Median payment invisible to the standard metric0.000497 ETH
Peak share of builder revenue unseen in one block72%
Builder revenue the metric missed overall18.5%

The conclusion went against my own project: the promising signal was an artifact of a blind instrument, and the opportunity was not real. I wrote that up, closed the finding, and shut the system down. I would rather kill my own thesis than ship a number I cannot defend — which is the same instinct a protocol is hiring when it hires a reviewer.

04 — Engagement

Available for contract and audit work

Best fit: DeFi protocol review, Solidity implementation, fork-test and PoC construction, and security work on lending, liquidation and cross-chain surfaces. Comfortable owning a scope end to end and reporting in writing.

There are also ten years of production full-stack and mobile engineering behind the security work — this is a specialisation, not a pivot away from shipping.

Start a Conversation contact@rebelstudiossoftware.com

Code samples, written reviews and proof-of-concept tests available on request.