Solana DEX Screener
Screener for early identification of Solana tokens that are pumping
Problem Statement
An Indian fund manager wanted an application that would screen tokens that trade on Solana DEXes in near real-time based on several criteria related to price action, tokenomics, liquidity and rug risk. The tool was required to scan the full universe of thousands of tokens including newly launched tokens and send alerts every hour based on user-defined screening rules.
Challenges
- Reliable data sources available for Solana DEXes were few in number, sparse in data availability and expensive
- Rate limiting prevented near real-time data extraction for thousands of tokens. An approach to identify tradable tokens from the decentralized universe was required.
- Designing of a framework that would allow for hundreds of possible screening rules
- Numerous disparate data sources increasing the data pipeline complexity
Solution
Summary
I developed and deployed a nextjs web application with various frontend features and a robust python backend. I established ETL pipeline processes for data management with routine cron-based scheduling.
Key Features
- Daily process to select tailored list of 400-500 active tokens based on data availability, liquidity, market cap and rug check
- Data pipeline to gather data from multiple sources mostly through REST APIs, then transform, compute, aggregate and store it
- Rules definition framework that allows for hundreds of possible rules to be applied to screen tokens based on price action, chart patterns, tokenomics, trade activity, market cap, liquidity, rug-check, etc.
- Intuitive user-friendly interface to setup and manage screeners
- Custom API for delivery of structured screener results that can be consumed by other applications
- Optional real-time telegram alerts
- Storage and querying of alerts history for analysis
- Authentication for secured access
Technologies Used
- React/Nextjs with typescript and tailwindcss for frontend
- NextAuth and bcrypt for authentication and encryption
- Python Flask backend
- PostgreSQL database with TimescaleDB and Materialized Views
- Python modules for ETL
- Cron for process scheduling
- Linux server with Nginx and systemctl for deployment
- SSL for security
Screener Rules Definition Framework
Screener Management Dashboard with Results
Screener Management Dashboard with Rules
Downloadable Alerts History
Results & Impact
Total tokens evaluated every day
Total tokens considered for screening
Screening frequency
Have a similar challenge?
Let's discuss how we can develop solutions for your specific use case.