Rutgers University New Brunswick, NJ, USA

Statistical Modeling, Stat Learning, Neural Networks, Machine Learning, Data Structures and Algorithms(Python), Database Management, Data Wrangling (R), NLP, Data Mining.

Andhra University Visakhapatnam, AP, India

Data Structures and Algorithms, Object-Oriented Programming, Matlab, Python.

EliteUS Software Solutions - Fedway Piscataway, NJ, US

• Developing an AI chatbot for Fedway system instructions using NLP and AI, integrating text and image data processing.

• Leveraged Language models from AWS Bedrock, Azure for AI deployment and Streamlit for dynamic interaction. Developed a Streamlit-based prototype chatbot integrated with Chroma, vector databases, utilizing PyMuPDF for PDF manipulation, Hugging Face Sentence Transformers for NLP embeddings.

• Deployed AI models using AWS Bedrock with LLaMA 3.1v for response generation and Pillow for image processing.

• Implemented a similarity matching algorithm using the CLIP model to compare text and image embeddings, utilizing NumPy for handling similarity matrix operations. Displayed matched text-image pairs, aligning image paths to corresponding textual instructions based on cosine similarity.

Harvest Software Solutions Jacksonville, FL, US

• Conducted real-time data extraction through APIs, retrieving information from Aniview. Managed and analyzed datasets exceeding 5 million rows, performing thorough exploratory data analysis (EDA) and incorporating Cost Per Mile (CPM) metrics for comprehensive performance evaluations.

• Executed data analysis using advanced visualization, summary statistics, and hypothesis testing to extract valuable insights into ad campaign performance.

• Developed predictive models, implementing XGBoost with a focus on minimizing Mean Squared Error (MSE). Applied a k-fold cross-validation approach to assess and mitigate overfitting risks. Leveraged H2O's AutoML for automated model selection, optimizing CPM predictions.

• Employed automation and advanced techniques, utilizing AutoML for tasks like feature engineering and hyperparameter tuning, ensuring the selection of optimal models for CPM prediction. Techniques applied include XGBoost, K-means, BERT, plotly, matplotlib, NLTK, spaCy, pandas, numpy, and ARIMA.

Rutgers University New Brunswick, NJ, US

• In my role as a teaching assistant for Statistics I, where we delved into essential concepts. I facilitated a deep exploration of Confidence Intervals and Testing Hypotheses, covering topics such as Population Proportion, Two Sample Inference, Chi-Square Tests (Goodness of Fit and Test for Association), ANOVA (Analysis of Variance), and provided a solid introduction to Regression.

• As a teaching assistant for 'Intro to Stat for Business,' I guided students through fundamental concepts such as Graphical Displays for Qualitative and Quantitative Data, Measures of Central Tendency, and Measures of Dispersion. We explored advanced topics like Bivariate Data, Combinatorics, Naïve Set Theory, Statistical Modelling and Basic Probability, providing a foundation for statistical analysis.

Accenture Sol. Pvt. Ltd. Hyderabad, TG, India

• Conducted client data analysis utilizing advanced querying techniques in Amazon Athena and Amazon Redshift. Implemented ETL ingestion in data pipeline for Regeneron, featuring data extraction, featuring data extraction methodologies that ensured optimal performance, data integrity, and seamless integration with downstream processes. Automated data pipelines with pandas, achieving a 40% increase in efficiency and 50% decrease in errors.

• Skillfully managed tasks and updates in JIRA, ensuring streamlined teamwork and prompt issue resolution. Streamlined data management processes within the data warehousing environment, implementing strategic DDL updates and deployment strategies that resulted in a substantial 15% improvement in system reliability.

• Proficiently crafted SQL queries using SQL Workbench for data manipulation and in-depth analysis. These queries were tailored to extract meaningful insights, optimizing data retrieval and providing a solid foundation for informed decision-making. Coordinated cross-team collaboration for deployments on production servers contributing to optimal system performance & functionality.

• Skills Obtained: Data Management, Data Warehousing, Advanced Data Visualization (Tableau), SQL, Problem Solving, Database Management, AWS (Amazon Web Services), Python (pandas).

Research Assistant Visakhapatnam, AP, India

• Implemented advanced computer vision techniques, to enhance the vehicle's perception and navigation capabilities. Object detection algorithms, including YOLO and Faster R-CNN, were utilized to accurately identify and locate various elements in the vehicle's surroundings, such as vehicles, pedestrians, traffic signs, and obstacles.

• Applied lane detection algorithms, by employing techniques like edge detection and Hough Transform, enables the vehicle to navigate within its lane securely. We employed image segmentation methods, such as Fully Convolutional Networks (FCNs), for partitioning images into semantic regions. Depth estimation techniques, utilizing stereo vision and monocular depth estimation, allowed the vehicle to gauge the 3D structure of its surroundings, crucial for assessing distances to obstacles.