StanQuant — Trading Algorithm Sandbox

StanQuant is a full-stack web platform for designing, uploading, and evaluating stock trading algorithms on millions of rows of historical data.

Built to empower both beginner and experienced traders, this system provides a secure and scalable environment to validate the logic and performance of your strategies before risking real capital.

---

Summary

As an undergraduate Computer Science student at the University of Waterloo, I created this project to solve a personal need — I wanted a place to test my own trading scripts using real, large-scale financial data with flexibility and safety.

This app lets users:

• Upload or create algorithmic trading strategies
• Evaluate them using detailed statistical backtesting tools
• Store, preview, and compare results for better decision-making

---

Tech Stack

• Frontend: Next.js + TailwindCSS
• Backend: Django + Django REST Framework
• Data Processing: Apache Spark (PySpark), Pandas, NumPy
• Database: AWS RDS (PostgreSQL) with SQLAlchemy ORM
• Script Execution: Kubernetes Jobs on AWS EKS with Dockerized workloads
• File Storage: Amazon S3 (managed via boto3)
• Security: All uploaded user files run inside isolated containers with timeouts, resource caps, and no external network access

---

Key Features

Algorithm Upload & Validation

• Upload .py scripts containing a single strategy function (def your_algo(df)).
• Must return a list or array of Buy/Sell/Hold signals matching the length of the DataFrame.
• Uploaded files are validated before execution.

Secure Execution Environment

• Each user-uploaded script is executed in a sealed Kubernetes container using AWS EKS.
• The job fetches historical stock data from AWS RDS, runs the algorithm, and stores output results back to S3.

High-Performance Data Handling

• Evaluations run on minute-level historical stock data (millions of rows per test).
• Uses PySpark and Pandas for fast, scalable processing of time-series indicators, rolling windows, and deciders.

Algorithm Evaluation System

• Users input metrics like:
  • Gain & Loss thresholds
  • Intercept Ranges
  • Clean conflict filtering
  • Position duration
• Algorithms can be combined and tested for consensus agreement.

---

Demo Video

To access a comprehensive walkthrough of the platform, view the Loom video demo:

Demo Video Folder on Loom

---

Photos

Evaluate Your Algorithms ↓

Evaluation Results ↓

View Past Evaluations ↓

Evaluation Analytics ↓

View Your Algorithms ↓

Preview Your Algorithms ↓

Delete Your Algorithms ↓

Upload Your Algorithms ↓

Create Your Algorithms ↓

Homepage ↓

Sign In ↓

Sign Out ↓

About Page ↓

User Guide ↓

---

Links

• Website: https://stanquant.com

Note: Backend deployment currently paused, please contact me.