Skip to content
All Systems
SYS_07 / 15SHIPPED • Alpaca Hackathon 2026
AI SYSTEMSPublic • TypeScriptMIT LicenseMulti-Agent AI & Deterministic Risk

Alpaca AI Trading Agents Hackathon (Aug–Sep 2026)

RazorStack Trading

Multi-Agent Quantitative Paper-Trading Platform

AI-powered quantitative paper-trading platform built for Alpaca AI Trading Agents Hackathon (Aug–Sep 2026), featuring a multi-agent research desk, deterministic risk engine, and real-time paper-trade execution.

Next.jsFastAPIPythonTypeScriptAlpaca Trading API

Architecture

M-01

Multi-Agent

Research Desk

Guardrails

M-02

Deterministic

Risk Engine

Execution

M-03

Real-Time

Alpaca Paper API

Backend

M-04

FastAPI + Next

Python + TypeScript

Section 01

Problem Statement

Autonomous LLM trading demos often place unconstrained orders without separating multi-agent research synthesis from deterministic position and risk limits.

Section 02

System Constraints

  • >Enforce deterministic risk checks prior to any broker order submission
  • >Coordinate multi-agent research signals with real-time Alpaca paper-trade execution

Section 03

System Architecture & Dataflow

Topology // 4 Nodes • 3 Edges
01

Multi-Agent Research Desk

Signals
02

Deterministic Risk Engine

Validated Orders
03

Alpaca Trading API

Fills & Positions
04

Next.js Trading Console

Section 04

Key Engineering Decisions