I'm a Data Scientist and Generative AI Practitioner passionate about building impactful solutions using NLP, Deep Learning, and Machine Learning. My work explores how AI can solve real-world problems—from sentiment analysis to Retrieval-Augmented Generation (RAG) systems.

• Fine-tuning and evaluating LLMs for custom NLP tasks
• Scaling RAG pipelines with LlamaIndex and semantic search
• Improving ML model fairness in imbalanced text classification
• Deploying real-time AI applications with Streamlit and Slack.
• Machine learning training using MLflow

• Parameter-Efficient Fine-Tuning (PEFT): QLoRA, LoRA, IA3
• Vector databases for context-aware search (e.g., FAISS, ChromaDB)
• Building production-ready GenAI apps and dashboards

• Applied NLP (especially underrepresented text domains), CV, ML, projects
• End-to-end MLOps/LLMOps pipelines
• AI/ML research with community impact

• Credit Card Fraud Detection Classifier - This project develops a robust machine learning pipeline for credit card fraud detection, leveraging a highly imbalanced dataset. Detecting fraudulent transactions is crucial for financial institutions to minimize monetary losses and protect customer assets.
• Resume & Job Application Advisor - An agentic system powered by CrewAI and open-source LLMs that helps job seekers improve their resumes, discover current job openings, generate tailored cover letters, and prepare for interviews — all in a single interactive Streamlit app.
• Racist Tweet Classifier — NLP pipeline with SymSpell correction, oversampling, stratified cross-validation & Streamlit UI
• Car Price Prediction — Regression modeling using XGBoost and feature engineering on real marketplace data
• RAG Toolkit — Experimenting with OpenAI, BGE, and LlamaIndex to build smarter retrieval pipelines
• Image Classifier - An End-to-End Image Classifier for 101 foodvision images capable of predicting the image label among the 101 foodvison image labels, e.g. sushi, pizza, etc. and a dog vision project in similar fashion using the 120 Stanford Dog breeds Image dataset

📩 LinkedIn • ✉️ chonzadaniel@yahoo.com • 🌐 GitHub Projects

💬 Ask me about NLP, Computer Vision, Machine Learning, Transformers, GenAI, fine-tuning LLMs, and practical AI deployment!