Chaithanya N

About Chaithanya N

Data Scientist with 5+ years of expertise in data extraction, pre-processing…

Experience

1. Data Scientist

   Citi

   Jan 2024 — Present · TX, US

   Engineered a custom LLM pipeline using LangChain to construct a knowledge graph from a vast scientific literature corpus, leading to a 30% improvement in information retrieval accuracy. Leveraged Google Cloud Platform services like Dataflow and Vertex AI to build and deploy efficient data pipelines and machine learning models, resulting in approximately 50% reduction in processing time and faster model training cycles. Performed complex data analysis on financial datasets, visualizing key metrics such as revenue growth, expenses, and profit margins using Tableau, resulting in a 25% reduction in report generation time. Led data analysis and predictive modeling initiatives using Python libraries (Scikit-learn, NumPy, pandas) and Spark (MLlib, PySpark), developed various advanced customer segmentation and behavior analysis algorithms, resulting in a 25% increase in customer lifetime value prediction accuracy. Successfully implemented Generative AI methods to overcome limitations in data availability for customer churn prediction, enabling the development of robust models. Utilized pre-trained models (BERT, GPT-2) within the Transformers framework to address Natural Language Understanding (NLU) and Natural Language Generation (NLG) tasks, accelerating project development and improving NLP capabilities. Implemented ML models (Random Forest, KNN, Naive Bayes models) to optimize loan approval processes, resulting in a 20% reduction in default rates and a 15% increase in approval efficiency, improving overall portfolio quality and profitability.