"Varun team on Alpaca AI Trading Agents Hackathon Hackathon"

Team Idea

Building Guardrail, a pre-trade risk layer for autonomous trading agents. The LLM proposes orders but cannot reach Alpaca directly. Every proposal is checked against limits defined in readable YAML (position size caps, drawdown halt, close-window block, symbol allowlist), and each decision is logged alongside the agent's stated rationale so you can see what it was trying to do when it got stopped. Python, alpaca-py, Anthropic tool use, FastAPI, paper trading only. Solo, not recruiting.

Team Members

Varun

varun_329

Submission

Guardrail

An autonomous AI options-trading agent that physically cannot reach the broker. The LLM only proposes a direction; a deterministic rules engine approves or blocks every order and logs why. Executes on Alpaca via its CLI. Paper trading.