Hojun Lee@hojundev

The Alpaca Quant Agent is an autonomous options-trading system that harvests the volatility risk premium, the persistent tendency for implied volatility to exceed realized volatility. Rather than buying options, it sells them, collecting premium through defined-risk credit spreads and iron condors on a curated, sector-diversified universe of 22 liquid ETFs and quality large-caps. It trades only when three filters agree: implied volatility is elevated (IV rank), the trend regime supports the structure (momentum, EMA crossover, ADX), and the name passes a quality and liquidity screen. Positions are sized with fractional Kelly and hard-capped at 2% of equity per trade. What keeps it safe is a wall of deterministic risk controls the AI cannot override: portfolio heat caps, a correlation-bucket concentration limit, a volatility term-structure exposure scaler that throttles down in stressed markets, an automatic delta hedge, and daily-loss and drawdown circuit breakers. The AI never invents trades. Deterministic code screens and risk-approves every candidate first. An LLM then checks live news, selects which approved trades to take, and writes the rationale, interacting with Alpaca's APIs through structured tools in the paper-trading environment. Autonomous where it adds value, bounded where it matters.

🌎 Online Hackathon | 💻 Dates: 28 August–4 September 2026 | 🏆 $6,000 Prize Pool Build AI trading agents on Alpaca — autonomous agents and trading apps using Alpaca's Trading API, MCP server and CLI. 🌍 Fully online hackathon. Join and build from anywhere in the world.