Hi Flashbots researchers and MEV searchers,
I’ve been developing an open-source Polygon arbitrage research framework focused on DEX price discrepancies, historical state reconstruction, atomic execution feasibility, and profitability analysis.
Rather than keeping the infrastructure private, I’ve decided to make the engineering work available for other researchers and builders to extend.
**GitHub:** GitHub - fasehab/polygon-arb-research · GitHub
### What’s implemented
- **Rust:** Blockchain monitoring, liquidity-pool discovery, pricing calculations, route optimization and preflight safeguards.
- **Solidity:** Balancer flash-loan execution contract prototype and local tests.
- **Python:** Historical transaction analysis, V2/V3 quote diagnostics and route discovery.
- **DEX research:** QuickSwap, SushiSwap, Uniswap V3 and Algebra liquidity pools.
- **Testing:** 29 passing Rust tests and successful optimized compilation in the latest verified source build.
### Where the research currently stands
The framework has demonstrated synthetic execution feasibility and historical analysis capabilities, but we have not established repeatable, net-profitable atomic arbitrage against real Polygon liquidity.
Several historical candidates were unprofitable after pool fees. Transaction-level V3 reconstruction and execution validation remain areas of active investigation.
I’m particularly interested in moving beyond isolated simulations toward empirically identifying profitable opportunities and understanding whether an independent searcher could realistically capture them.
### Looking for collaboration on
1. Transaction-level historical MEV opportunity discovery.
2. Efficient multi-venue and multi-pool route optimization.
3. Low-latency Polygon infrastructure and transaction inclusion.
4. Atomic execution, gas-aware profitability and failed-execution analysis.
5. Independent code review and reproducible research.
An interesting research question is how much of the observed arbitrage spread is realistically capturable once liquidity, inclusion competition, latency and execution costs are considered.
**This is research infrastructure, not a production-ready or proven profitable trading bot.**
If you’re working on similar problems, I’d welcome technical feedback, reproducible experiments, GitHub contributions or collaboration.
**Repository:** GitHub - fasehab/polygon-arb-research · GitHub
Looking forward to discussing the engineering and economics with the MEV research community.