QuantumEdge
built options pricing and ETF arbitrage models to place in the top 1% globally in the IMC Prosperity challenge out of 22,000+ teams.
imc prosperity is this massive global quant trading competition. there were 22,000 teams competing, mostly cs graduates, math PhDs, and actual traders from top-tier universities. my friends tanish, pavitra, and i decided to enter because we liked math and wanted to see if we could build a trading engine that wouldn't immediately crash under simulated market conditions. we had absolutely zero prior quant experience when we started. we basically spent the first week doing nothing but reading papers on market microstructure, cointegration, and option Greeks. we ended up designing three separate trading tracks. first was an ETF cross-product arbitrage model that used GARCH volatility models and Johansen cointegration tests to spot tiny price discrepancies across correlated assets. second was a complete options valuation engine written in Python (about 3,000 lines) that inverted Black-Scholes to calculate implied volatility and managed risk by balancing delta and gamma exposures in real-time. finally, we built a market-making bot that placed bid/ask orders based on order book depth-vanishing signals, optimizing parameters using Hyperopt TPE. we were coding between school classes and staying up until 4 AM trying to fix latency issues. ending up in the top 1% globally was crazy, but the coolest part was realizing that quantitative finance isn't some mystical secret—it is just math, clean data pipelines, and a lot of testing.