Venkat Chandarlapati

Modular Python pipeline built with OOP design and the Observer pattern — dedicated classes for data ingestion, processing, and report generation trigger downstream updates automatically. Reduced processing errors from 12% to under 2% through systematic edge-case testing and input validation. Full codebase managed on GitHub with structured branching and inline documentation across the SDLC.

• Python
• Observer Pattern
• Pandas / NumPy
• Data Pipeline

Core business logic rebuilt in Java using OOP (encapsulation, inheritance) and the Factory pattern to model employee entities, with MVC separation decoupling data access from reporting. Analyzed 1,000+ records with multi-table JOINs, GROUP BY, and aggregates; improved query performance by 20% through schema normalization and strategic indexing.

• Java
• SQL
• Factory Pattern
• MVC
• Query Optimization

Real-time gesture-recognition system built from scratch — improved detection accuracy from a 60% baseline to 85%+ through iterative algorithm tuning and test-driven improvement cycles. Multi-threaded frame processing cut input lag by roughly 40%, applying concurrency to achieve responsive real-time performance.

• Python
• OpenCV
• Multithreading
• Real-time CV

Image classification system using a convolutional neural network to detect and categorize skin conditions from input images — covers the full pipeline from image preprocessing through model inference.

• Python
• CNN
• Image Classification
• Deep Learning