Projects Portfolio¶

Featured Projects¶

OpenSAMPL - Open Source Clock Probe Aggregator and Visualizer¶

Advanced clock synchronization and timing analysis platform

A comprehensive open-source platform for clock probe aggregation and visualization, developed at Oak Ridge National Laboratory. OpenSAMPL provides advanced tools for precise clock monitoring, time synchronization analysis, and timing accuracy verification across distributed systems.

Technologies: Python, NumPy, SciPy, Matplotlib, Pandas, TimeSync, Network Time Protocol (NTP), Docker

Key Features: - Clock Probe Aggregation: Centralized collection and analysis of clock signals - Time Synchronization Analysis: Precise measurement of clock drift and synchronization - Visualization Tools: Interactive dashboards for timing analysis and monitoring - Distributed Systems Support: Multi-node clock monitoring and analysis - Performance Optimization: High-precision timing algorithms and measurements - Research Applications: Scientific computing, real-time systems, embedded development

Repository: github.com/ORNL/OpenSAMPL Documentation: ornl.github.com/OpenSAMPL PyPI Package: pypi.org/project/opensampl

BlackLake - S3-based Data Portal¶

Enterprise-grade data artifact management platform

A production-ready, enterprise-grade data artifact management platform that combines modern technology with comprehensive features for data management, search, governance, and compliance. Built with Rust and featuring a complete S3-based Git CLI and application stack for data.

Technologies: Rust, Axum, React, TypeScript, PostgreSQL, Apache Solr, Redis, MinIO, Docker, Prometheus, Grafana, OpenTelemetry

Key Features: - Multi-Tenant Architecture: ABAC policies with tenant isolation - Advanced Security: OIDC/JWT authentication, RBAC, audit trails - Comprehensive Governance: Branch protection, quotas, retention policies - Production Operations: Monitoring, backup, disaster recovery - Modern UI: React interface with mobile support and PWA capabilities - Developer Experience: CLI tools, SDKs, comprehensive documentation

Live Demo: BlackLake Documentation Repository: s3-rust-data-portal

Final Fantasy Football¶

A semantic learning algorithm for fantasy football

A sophisticated machine learning system that applies semantic analysis to fantasy football decision-making. The project combines web scraping, data processing, and predictive modeling to provide actionable insights for fantasy football players.

Technologies: Python, scikit-learn, Pandas, NumPy, Matplotlib, Requests, BeautifulSoup, Flask, Docker, Compose, Airflow

Key Features: - Automated data collection from multiple sources - Semantic analysis of player performance patterns - Predictive modeling for roster decisions - Web interface for user interaction - Containerized deployment with orchestration

Where I've Been¶

A web application to visualize travel history

An interactive web application that allows users to visualize and explore their travel history through an intuitive mapping interface. The application processes location data and presents it in an engaging, interactive format.

Technologies: Python, Flask, Docker, Compose, PostgreSQL, Leaflet, Vue, Bootstrap

Key Features: - Interactive map visualization using Leaflet - Travel timeline and statistics - Data import/export capabilities - Responsive design with modern UI - Secure user authentication

This Is A Casino¶

Trade/visualize stocks via semantic, RL, DL

A comprehensive stock trading platform that leverages semantic analysis, reinforcement learning, and deep learning techniques to provide intelligent trading insights and portfolio management.

Technologies: Python, Flask, Docker, Compose, PostgreSQL, Leaflet, React, Bootstrap

Key Features: - Real-time market data processing - Semantic analysis of market sentiment - Reinforcement learning for trading strategies - Interactive portfolio visualization - Risk assessment and management tools

Decentralized Content Reward System¶

A decentralized content reward system for the web

An innovative Web3 platform that enables content creators to monetize their work through a decentralized reward system, leveraging blockchain technology for transparent and fair compensation.

Technologies: Python, Rust, Flask, Docker, Compose, PostgreSQL, Leaflet, React, Bootstrap

Key Features: - Blockchain-based reward distribution - Smart contract integration - Content verification system - User reputation management - Decentralized governance

Genesis¶

Audio-input game to create a universe

An experimental game that uses audio input to procedurally generate and evolve a virtual universe. Players interact with the system through sound, creating unique and dynamic worlds.

Technologies: Python, pygame

Key Features: - Real-time audio processing - Procedural universe generation - Interactive sound-based controls - Dynamic visual effects - Save/load universe states

Cosmic Architect¶

Compete to build the best planet

A competitive game where players design and build planets, competing against others to create the most successful planetary ecosystem.

Technologies: Python, pygame

Key Features: - Planet design interface - Ecosystem simulation - Multiplayer competition - Scoring and ranking system - Real-time gameplay mechanics

pygarden¶

Python package for geospatial data processing and analysis

A comprehensive Python package for geospatial data processing, spatial analysis, and machine learning applications. Developed as part of the OpenSAMPL ecosystem at Oak Ridge National Laboratory.

Technologies: Python, NumPy, Pandas, GeoPandas, Shapely, Rasterio, GDAL, scikit-learn

Key Features: - Spatial Data Processing: Efficient handling of vector and raster data - Machine Learning Integration: Spatial ML algorithms and workflows - Data Format Support: Multiple geospatial data format support - Performance Optimization: High-performance spatial operations - Research Tools: Advanced spatial analysis capabilities - Documentation: Comprehensive API documentation and examples

PyPI Package: pypi.org/project/pygarden Repository: github.com/ORNL/pygarden

maw¶

CLI for quickly combining tabular data into Parquet

A high-performance command-line tool written in Rust for efficiently combining and processing tabular data into Parquet format, optimized for speed and memory usage.

Technologies: Rust

Key Features: - High-performance data processing - Memory-efficient operations - Command-line interface - Support for multiple input formats - Optimized Parquet output

Project Categories¶

Research & Open Source Platforms¶

• OpenSAMPL: Open source clock probe aggregator and visualizer
• pygarden: Python framework for data processing and fast project setup

Enterprise Data Platforms¶

• BlackLake: Production-ready data artifact management platform
• S3-based Git CLI: Command-line interface for data version control

Data Engineering & Analytics¶

• Final Fantasy Football: Machine learning and data processing
• maw: High-performance data processing tool

Web Applications¶

• Where I've Been: Travel visualization platform
• This Is A Casino: Financial trading platform
• Decentralized Content Reward System: Web3 content platform

Interactive Games¶

• Genesis: Audio-based universe creation
• Cosmic Architect: Planet building competition

Technical Approach¶

All projects demonstrate a focus on:

• Scalable Architecture: Containerized deployments with Docker
• Modern Web Technologies: React, Vue, and responsive design
• Data Processing: Efficient handling of large datasets
• User Experience: Intuitive interfaces and interactive visualizations
• Performance: Optimized code and efficient algorithms

Open Source Contributions¶

GitHub Profile¶

Profile: github.com/sempervent

All projects are available on GitHub and demonstrate: - Clean, well-documented code with comprehensive README files - Comprehensive testing with automated CI/CD pipelines - Docker containerization for consistent deployment - API documentation with OpenAPI/Swagger specifications - User guides and examples for easy adoption

Key Repositories¶

• s3-rust-data-portal: Enterprise data management platform
• Data Engineering Tools: High-performance CLI utilities and data processing libraries
• Web Applications: Full-stack applications with modern frameworks
• Machine Learning Projects: ML pipelines and predictive modeling systems

Contribution Philosophy¶

• Open Source First: All personal projects are open source
• Documentation Driven: Comprehensive documentation for all projects
• Community Focused: Welcoming contributions and feedback
• Production Ready: Enterprise-grade code quality and testing

Future Projects¶

Advanced Geospatial Data Processing¶

• Real-time IoT Data Streams: Processing and analyzing sensor data from IoT devices
• Spatial Machine Learning: Applying ML techniques to geospatial datasets
• Cloud-native Geospatial Architectures: Scalable solutions for large-scale spatial data

Enterprise Data Management¶

• Data Mesh Architecture: Implementing domain-driven data architecture patterns
• Advanced Data Governance: Enhanced compliance and data lineage tracking
• Federated Data Platforms: Cross-organization data sharing and collaboration

AI/ML Integration¶

• MLOps Pipeline: End-to-end machine learning operations
• Semantic Search Enhancement: Advanced natural language processing for data discovery
• Automated Data Quality: AI-powered data validation and cleansing

Performance & Scalability¶

• Edge Computing: Distributed processing for real-time applications
• Multi-cloud Strategies: Hybrid cloud data management
• Advanced Caching: Intelligent data caching and optimization

Developer Experience¶

• Enhanced SDKs: Improved developer tools and APIs
• Visualization Tools: Interactive data exploration interfaces
• Documentation Automation: AI-powered documentation generation