randika@portfolio:~$ echo "Ready to innovate! 🚀"█
|
|
name: M. A. Randika Prabashwara
role: Data Science Engineer & Research Assistant
education: B.Sc. Engineering (Honours) - Computer Science & Engineering
specialization: Data Science & Engineering
university: University of Moratuwa, Sri Lanka
research_focus: AI-ML Solutions, Computer Vision, Super Resolution
status: Recent Graduate Student & Research Assistant
location: Sri Lanka
experience: 3+ years in ML/AI development- 🔬 Research: Advanced Computer Vision, Super Resolution & AI-ML Solutions
- 💼 Work: Building enterprise-grade data solutions and intelligent platforms
- 🎓 Study: Final year at University of Moratuwa (Data Science & Engineering)
- 🏆 Achieve: Top 83% Kaggler with proven ML competition track record
- 📚 Mentor: Teaching Assistant supporting 100+ students in programming & compiler design
🎓 Final Year Research Project | Jun 2024 - Present
Revolutionary super-resolution framework for scientific imaging using conditional structure-informed multi-scale learning with advanced encoder-based feature integration and custom Semantic Structural Loss (SSL).
🎯 Impact & Results:
- ✅ 3.5dB PSNR improvement, 20% SSIM increase, 20% LPIPS decrease over state-of-the-art methods
- ✅ Outperformed EDSR, ESRGAN, SwinIR & HMANet on SEM dataset & benchmark datasets (Set5, Set14, BSD100)
- ✅ Novel architecture with encoder-driven conditioning & three-stage pipeline optimization
- ✅ Published at ADScAI Summit 2025 - Applied Data Science & Artificial Intelligence
🛠️ Tech Stack: Python TensorFlow Keras PyTorch CUDA OpenCV Computer Vision Deep Learning
🏢 Data Science Engineer Intern | Dec 2023 - Jun 2024
Enterprise-grade sales analytics platform with real-time GPS tracking, behavioral analysis, and automated report generation serving field teams across Sri Lanka.
🎯 Core Capabilities:
- 🗺️ Real-Time GPS Tracking - Live location monitoring with territory coverage analysis & route optimization
- 🤖 ML-Powered Intelligence - Sales forecasting, performance scorecards & opportunity identification
- 📊 Comprehensive Dashboards - Executive analytics with dynamic visualization & automated reporting
- ⚡ 35% Efficiency Improvement - Reduced operational costs through predictive analytics integration
🛠️ Tech Stack: Python FastAPI Apache Kafka MySQL Redis React Docker AWS Machine Learning
🤖 Computer Vision Solution | Apr 2025 - May 2025
Revolutionary multi-modal OCR pipeline combining TrOCR, EasyOCR, and Enhanced Tesseract with intelligent multi-factor orientation detection for Sri Lankan driving license processing.
🎯 Key Achievements:
- ✅ 96% accuracy in vehicle class extraction, 92% accuracy in data extraction
- ✅ 1.8 seconds average processing time with 98% orientation detection success
- ✅ Adaptive image enhancement - No segmentation dependencies, handles complex layouts
- ✅ Production-ready pipeline - Deployed with comprehensive error handling
🛠️ Tech Stack: TrOCR EasyOCR PyTorch OpenCV Transformers CUDA Computer Vision
💡 Click to expand more Featured Projects
🎼 Big Data ML Pipeline | Mar 2025 - Jun 2025
Scalable machine learning pipeline built on Apache Spark for distributed music genre classification with ensemble approach combining Random Forest, Gradient Boosting Trees, and Logistic Regression.
🎯 Core Capabilities:
- 🔥 Distributed Training - Implemented scalable feature extraction and distributed clusters
- ⚡ Ensemble Learning - Multi-algorithm approach with automated model selection
- 📊 Production Pipeline - Complete MLOps workflow with model persistence & deployment
- 🎯 High Performance - Optimized for large-scale audio processing workloads
🛠️ Tech Stack: Python Apache Spark PySpark MLlib Audio Processing Distributed Computing
🤖 Autonomous Driving Research | Jan 2025 - Jun 2025
Hybrid framework combining transformer-based neural networks with Bayesian filtering for uncertainty quantification and safety mechanisms in autonomous driving scenarios.
🎯 Innovation Focus:
- 🧠 Hybrid Architecture - Neural networks with physics-informed constraints
- 🎯 Uncertainty Quantification - Bayesian filtering for safety-critical predictions
- 🚗 Real-World Application - Designed for autonomous driving trajectory planning
- 📊 Performance Metrics - Enhanced prediction accuracy with safety guarantees
🛠️ Tech Stack: PyTorch TensorFlow Transformers FMLC FilterPy CARLA
🏦 Financial Document Processing | Previous Project
Intelligent cheque processing system with collaborative 3+ member development team eliminating manual data entry bottlenecks and integrating with Google Cloud Vision API and Azure Computer Vision API.
🎯 Key Features:
- 💡 Automated Processing - End-to-end cheque data extraction pipeline
- 🔗 Multi-API Integration - Google Cloud Vision & Azure Computer Vision
- 👥 Team Collaboration - 3+ member development with custom vision models
- 📱 Mobile Integration - Seamless mobile application integration
🛠️ Tech Stack: Python Google Cloud Vision API Azure Computer Vision API OpenCV Mobile Integration
🔥 Click to expand my technical arsenal
Programming Languages: 
AI/ML & Data Science: 
Web Development: 
Databases: 
Big Data & Streaming: 
Cloud & DevOps: 
Tools & Others: 
---
🎓 Academic & Professional Achievements
AMP®-Parkinson's Disease Progression Prediction - Achieved top 83% ranking among global participants
- 🥇 Best Result in Physical Science Stream - Prince of Wales College, Moratuwa
- 🏆 Mathematical Olympiad High Distinction (2011-2017) - Consistent high performance
- 🎓 B.Sc. Engineering (Honours) - University of Moratuwa (Final Year)
- 📝 Published Research - ADScAI Summit 2025 (Applied Data Science & Artificial Intelligence)
- 🏅 Island Rank 29 - G.C.E Advanced Level (Combined Mathematics, Physics, Chemistry)
- ☁️ AWS Academy Graduate - Data Engineering, Machine Learning, NLP
- 🎓 Machine Learning Specialization - Stanford University (Coursera)
- 🔬 Operations Research - National Taiwan University
- 🧠 Mathematics for Machine Learning & Data Science - DeepLearning.AI
- 🏥 AI for Medical Diagnosis - DeepLearning.AI
- 🔬 Research Assistant - University of Moratuwa (Jul 2025 - Present)
- 👨🏫 Teaching Assistant - University of Moratuwa (Jan 2025 - Jun 2025)
- 💼 Data Science Engineer Intern - Pinnacle Company Ltd (Dec 2023 - Jun 2024)
mindmap
root((Research Focus))
Advanced Computer Vision
Super Resolution
Scientific Imaging
Structure-Informed Learning
Deep Learning Architectures
Transformer Networks
Generative Adversarial Networks
Multi-Scale Learning
AI for Autonomous Systems
Trajectory Prediction
Uncertainty Quantification
Safety-Critical AI
Enterprise Solutions
Real-time Analytics
Geospatial Intelligence
MLOps & Deployment
Research Publications
Academic Writing
Conference Presentations
Python 15 hrs 45 mins ████████████████████░ 85.2%
JavaScript 1 hr 30 mins ██░░░░░░░░░░░░░░░░░░░ 8.1%
SQL 45 mins █░░░░░░░░░░░░░░░░░░░░ 4.1%
CUDA 25 mins ▒░░░░░░░░░░░░░░░░░░░░ 2.3%
Others 5 mins ░░░░░░░░░░░░░░░░░░░░░ 0.3%
🚀 Open to exciting opportunities in: 
"Seeking opportunities to apply advanced technical and analytical skills to address complex real-world problems and deliver impactful AI-driven solutions for industry transformation."
