PUMP — Trading & Analytics Platform

The production application is part of a commercial, paywalled analytics suite, but you can get a good sense of its positioning and UI style from the public promo page: https://zenpulsar.com/solutions

Overview

PUMP is a sentiment analytics platform for traders that aggregates posts from multiple social media sources (Reddit, Twitter, Telegram, Weibo, Guba, and others), applies ML-based sentiment prediction to each post, and presents the data through interactive visualizations. The platform tracks 5,000+ assets, primarily focused on cryptocurrency markets.

This was my first valuable commercial project — I joined a small team after they secured investor interest with an initial MVP and stayed for over two years, growing from frontend developer to full-stack engineer.

My Role

Full-Stack Developer (Backend-Heavy) — Started as frontend developer but quickly transitioned to full-stack, with the majority of my work focused on backend systems.

• May 2022: Joined as Vue frontend developer to reduce load on the full-stack developer
• Sep 2022: Became full-stack developer — immediately started maintaining and updating backend endpoints, performing refactoring, optimizations, and general improvements
• May 2023: Took full ownership of all BFF (Backend-for-Frontend) endpoints after a developer left the company
• 2023: Mentored a junior developer for 2-3 months — trained him and conducted code reviews
• Maintained and improved social media parsers and ETL pipelines in Airflow
• Worked across multiple cloud providers as the team migrated: AWS → GCP → Azure

While I owned the frontend entirely, the bulk of my contributions were on the backend — API development, data pipelines, database optimizations, and parser maintenance.

Key Features

Charts & Analytics

• Post volume charts with filters for popularity metrics
• Time-series visualization of sentiment trends per asset
• Lightweight Charts integration for trading-style interactions

Treemap

• Visual representation of asset sentiment distribution
• Quick overview of market sentiment across 5k+ assets

Search

• Weaviate-powered semantic search across MongoDB post storage
• Filter and explore millions of social media posts

Graph Visualization

• Neo4j-based relationship mapping between posts and reposts
• Visualize how content spreads and which accounts drive engagement

Spider Charts

• Multi-metric visualization per post using Chart.js
• Metrics normalized to 0-1 scale for comparison

RAG Chatbot

• Research assistant added in late 2023
• Context-aware responses based on stored social media data

Technical Implementation

Backend

• Python / FastAPI for REST APIs
• MongoDB for post storage
• Neo4j for graph relationships
• Weaviate for semantic search

Frontend

• Vue 2 → Vue 3 migration (completed in 2023)
• Vuetify component library
• Virtual scroll implementations for large datasets
• Lightweight Charts for trading-style visualizations

Data Pipeline

• Airflow for ETL orchestration
• Custom parsers for Reddit, Twitter, Telegram, Weibo, Guba, and other sources
• Refactored parser architecture with SOLID principles (early 2023)

Cloud & Storage

• Migrated across AWS → GCP → Azure due to investment constraints
• Worked with S3, BigQuery, Google Blob Storage, Azure Blob Storage
• Parquet files for efficient data storage and transfer

Timeline

Tech Stack

Python

Vue

FastAPI

MongoDB

Airflow

Lightweight Charts

Chart.js

OpenAI

Pandas

Redis

AWS

GCP

Azure

• Backend: Python, FastAPI, MongoDB, Neo4j, Weaviate
• Frontend: Vue 3, Vuetify, Lightweight Charts, Chart.js
• Data Pipeline: Airflow, Pandas, Parquet
• Cloud: AWS, GCP, Azure (S3, BigQuery, Blob Storage)