Data Scientist in Generative AI & NLP

Sai Gayathri Reddy Bhimireddy

Edison, New Jersey, United States ********************@*****.*** 716-***-**** linkedin.com/in/saigayathri20 PROFESSIONAL SUMMARY

Data Scientist with 4+ years of experience developing and deploying end-to-end AI/ML solutions — including Generative AI, Large Language Models, deep learning, and classical machine learning — to solve complex business problems in regulated environments. Proficient in Python, SQL, and R with expertise in the full ML lifecycle: data pipeline design, feature engineering, model selection, hyperparameter optimization, statistical validation, and production deployment. Experienced applying NLP and transformer-based architectures, agentic system concepts, and Responsible AI practices including model documentation, explainability, and bias evaluation. Strong collaborator — skilled translating complex technical concepts and data science outputs into clear, actionable insights for cross-functional product, engineering, and business partners. MS in Data Science, University at Buffalo. EDUCATION

University at Buffalo, State University of New York Aug 2023 - Jan 2025 Master of Science, Data Science

SKILLS

Programming Languages: Python, SQL, R, PySpark

Frameworks & Libraries: TensorFlow, Keras, Scikit-learn, Pandas, NumPy, Hugging Face Transformers, LangChain, LangGraph Databases: Databricks, Snowflake, Amazon Redshift

Tools & Software: Git, GitHub, Streamlit, Tableau, Power BI, Matplotlib, Seaborn Cloud Platforms: AWS, Azure

Soft Skills: Data storytelling, Executive-level insight delivery, Technical concept translation, Stakeholder-facing communication, Collaboration

WORK EXPERIENCE

Camber Pharmaceuticals, Inc. Data Scientist Feb 2025 - Present

• Designed, developed, and deployed an end-to-end GenAI/NLP solution using transformer-based LLM architectures (Hugging Face Transformers), achieving ~88% accuracy; delivered data analysis outputs, maintained model documentation and Responsible AI evaluation, and integrated the solution into operational workflows.

• Built scalable ML data pipelines in Python and SQL processing 100K+ records; applied feature engineering, model selection, A/B testing, and hyperparameter optimization to maximize predictive accuracy, and monitored post-deployment model performance for drift and quality maintenance.

• Applied statistical trending tools, hypothesis testing, and EDA on complex pharmaceutical datasets to identify trends and patterns; produced Tableau and Power BI visualizations translating ML outputs into actionable insights for cross-functional and executive audiences.

• Partnered with product, engineering, and business stakeholders to translate ambiguous requirements into technical AI/ML specifications; applied multivariate statistical analysis to prioritize drivers, automated analytics workflows reducing manual effort by 30%, and improved decision-making speed.

Amrita Vishwa Vidyapeetham Research Data Scientist (AI/ML) – Deep Learning Research Project Jun 2022 - May 2023

• Designed and built a production-grade deep learning CNN model (94% accuracy, 12,000+ samples) applying best practices in model selection, hyperparameter optimization, statistical validation, bias evaluation, and Responsible AI documentation; optimized inference performance to 30+ FPS and deployed to a production interface.

• Implemented version-controlled Python ML pipelines and reusable components; used data analytics software to communicate experiment results via clear visualizations, and mentored junior researchers on ML methodologies, experimental design, and data visualization, co-authored peer-reviewed research (IPACT 2023). Hetero Labs Pvt Ltd Data Scientist Mar 2020 - Apr 2022

• Built scalable ETL data pipelines in Python and SQL processing 1M+ records; performed statistical analysis and predictive modeling

(classification and regression) to surface operational bottlenecks, reducing delays by 15% and informing cross-functional business planning.

PROJECTS

Customer Churn Prediction — End-to-End AI/ML Production System

• Architected and deployed a production-grade AI/ML system on 10K+ records (Python, Scikit-learn): data pipelines, feature engineering, A/B testing, model training, bias assessment, and explainability outputs (85–88% accuracy); deployed a live Streamlit application with real-time scoring, KPI monitoring, and model performance tracking. Brain Tumor Classification (CNN) & Loan Approval Prediction

• Developed deep learning and classical ML models (94% accuracy; 35% improvement) using Python, TensorFlow, and Scikit-learn; implemented model documentation, Responsible AI validation, reusable ML components, and production interfaces for real-time inference.

CERTIFICATIONS & ACHIEVEMENTS

• Google Data Analytics Certificate - Google

Contact this candidate