Data Scientist Machine Learning

Tanvi Patel

Data Scientist

Jersey City, NJ *****.*.*******@*****.*** +1-201-***-**** GitHub: Tanvi3004 LinkedIn PROFESSIONAL SUMMARY

Data Scientist with 4+ years of experience designing and deploying scalable, data-driven solutions across diverse industries. Proficient in machine learning, deep learning, NLP, and cloud computing, with expertise in regression, classification, clustering, transformer models, attention mechanisms, encoder-decoder architectures, LSTMs, and CNNs. Skilled in Python, R, SQL, TensorFlow, Keras, PyTorch, PySpark, and data analytics libraries such as Pandas, NumPy, and Matplotlib. Experienced in cloud-native development, deploying models via AWS, Azure, and container orchestration tools like Kubernetes and Terraform. Strong background in Big Data processing (Spark MLlib, PySpark, Redis, MongoDB, PostgreSQL). Proficient in Natural Language Processing (NLP) for text classification, sentiment analysis, and document summarization. Knowledgeable in Generative AI and Large Language Models (LLMs), including Hugging Face Transformers, LangChain, fine-tuning, and model deployment. WORK EXPERIENCE

Data Scientist Principal Financial Group, NJ Jan 2024 - Present

● Led end-to-end AI/ML model development using Python, SQL, and Spark, optimizing predictive analytics and business intelligence workflows.

● Designed and deployed fraud detection models using LSTM, BERT, and Random Forest, increasing anomaly detection accuracy by 30%.

● Developed ML pipelines for feature engineering and hyperparameter tuning, resulting in a 15% improvement in model performance.

● Built scalable ETL pipelines with Pandas and Spark, reducing data preprocessing time and enhancing training efficiency.

● Designed and trained encoder-decoder architectures with attention mechanisms for NLP tasks, improving text summarization and document translation models.

● Developed time-series forecasting models (LSTMs, BiLSTMs) for financial trend analysis, improving prediction accuracy by 30%.

● Deployed MLOps pipelines using AWS SageMaker, Docker, and Terraform, reducing model deployment time by 40%.

● Integrated LLMs and NLP models for internal knowledge retrieval and automation of document workflows. Data Scientist JKSOL Infotech, India Mar 2018 –Jul 2021

● Applied Agile methodologies (Scrum, Kanban) to streamline ML model development, from data collection to deployment.

● Built interactive dashboards using Tableau and Pandas, providing real-time insights and increasing project ROI by 20%.

● Developed machine learning models (Random Forest, SVM, XGBoost, Gradient Boosting) for customer segmentation and predictive analytics, boosting decision-making efficiency by 25%.

● Implemented sentiment analysis models using TF-IDF, Word2Vec, and LSTMs, improving brand perception analytics for marketing teams.

● Optimized scalable data pipelines using AWS (EC2, S3, Glue), reducing data processing time by 40%.

● Designed classification models (logistic regression, SVM, neural networks) to enhance predictive analytics and business intelligence.

● Researched and implemented advancements in Seq2Seq architectures and BiLSTMs for speech recognition and text classification, integrating attention mechanisms for accuracy improvements. SKILLS

Methodologies : SDLC (Scrum, Agile, Kanban)

Languages: & IDE’s : Python, R, DAX, Visual Studio Code, PyCharm, Jupyter Notebook Frameworks & Language : Anaconda, TensorFlow, Python, Keras, PyTorch, Hugging Face Transformers Data Analysis and Visualization : Pandas, NumPy, Matplotlib, ggplot2, SciPy, Keras, Scikit learn, OpenCV, Tableau, Power-BI Databases : MySQL, PostgreSQL, RDBMS, NoSQL (REDIS, MongoDB, Elastic Cache, DynamoDB) Machine Learning & AI : Deep Learning, NLP, Transformer Models, Neural Network, PySpark Cloud Technologies : AWS, (EC2, S3, Glue, AWS Athena and AWS Quick sight) Azure, Snowflake Generative AI & LLMs : GPT-based models, Hugging Face, LangChain, Text Generation, Summarization, RAG) Developer & Other Tools : Glue ETL, Athena, Kubernetes, Sagemaker, Terraform, Databricks, Git and GitHub EDUCATION

Master of Science in Data Science – New York Institute of Technology, Old Westbury, NY - USA Master of Science in Applied Mathematics – Parul University, Gujarat - IND B.E in Electronics and Communication Engineering – Gujarat Technological University (GTU), Gujarat – IND, CERTIFICATIONS & PROJECT PORTFOLIO

The Path to Insights: Data Models and Pipelines, The Power of Statistics, google business intelligence, google advanced data analytics, Go Beyond the Numbers: Translate Data into Insights, The complete SQL, Complete Data Science, Machine Learning, Deep learning Foundations of Data Science

RESEARCH PUBLICATIONS

“Analysis of a Prey–Predator Model"

Published by Taylor & Francis Group as part of the book chapter Mathematical Modeling and Soft Computing in Science and Engineering.

Co-authored research on prey–predator dynamics using differential equations and simulations to validate ecological stability.

Publication Link: Taylor & Francis - Analysis of a Prey–Predator Model

Contact this candidate