I'm a recent summa cum laude IT graduate from Kennesaw State University. I recently built a machine learning classifier model from scratch using the python libraries NLTK and Scikit-Learn for NLP and model training to predict user sentiment from text without relying on HuggingFace transformers. I used 42,000+ user feedback dataset entries and achieved a 56% accuracy rate for emotion detection in the first 2 weeks of the project. This was part of an end-to-end Customer Support AI application I built with Python, React/TypeScript, Azure, and ML/LLM models that processed 1,000+ customer feedback entries and automated 80% of the manual response work. I'm a problem-solver with proven ability to analyze systems, secure infrastructure, design data pipelines, and support users.

I have hands-on experience in full-stack development, machine learning, cloud systems, database work, cybersecurity, and technical support.

Find my work below.

Overview
CapSense AI is a full-stack system for enterprise customer feedback: sentiment, emotion, sarcasm, aspect analysis, plus empathetic responses via Phi-3 Mini. Built with React, Flask, and Azure. Supports single input and CSV batch processing.

Key features

• Sentiment classification (positive / neutral / negative)
• Emotion detection (anger, joy, anticipation, neutral, disgust, sadness)
• Sarcasm detection
• Aspect-based sentiment analysis
• Empathetic AI responses (Phi-3 Mini)
• CSV batch processing + Sentiment Analysis Report UI
• Azure App Services + SQL DB integration

Tech stack

• Frontend: React 18, Vite, Bootstrap, Chart.js
• Backend: Python 3.9+, Flask, PyODBC, joblib, NLTK
• ML: Naive Bayes (emotion), Hugging Face transformers (sarcasm)
• Cloud & CI: Azure AI Studio (Phi-3), Azure App Services, GitHub Actions

Highlights

• Naive Bayes emotion classifier trained on 42k+ samples
• Modular React UI with report components for each detector
• Robust backend endpoints for individual & batch analysis
• Edge-case handling for empty/malformed/large CSVs
• Production deployment and Azure optimizations

Repository: insert your repo link

Movie SPA — single page app using The Movie Database API. Search, filter, paginate, show cast + details. Built with jQuery, AJAX, JSON & REST.

Languages: Python · JavaScript · TypeScript · SQL · C++ · Java
Libraries / Frameworks: React · Flask · Chart.js · scikit-learn · Hugging Face · NLTK
DB / Cloud: SQL Server · Azure · GitHub Actions
Tools: Postman · pytest · VS Code

[![LinkedIn](https://img.shields.io/badge/LinkedIn-Connect-blue?logo=linkedin)](https://www.linkedin.com/in/vkwenda)
[![Email](https://img.shields.io/badge/Email-Contact-red?logo=gmail)](mailto:vkwenda@students.kennesaw.edu)
[![Resume](https://img.shields.io/badge/Resume-Download-green?logo=google-docs)](https://your-resume-link.com)