Vineeth Reddy Thumma

Amperity (Customer Data Platform):

(Databricks, SQL, Spark SQL, Tableau, PPT, Excel)

• Eddie Bauer Kids' Shoppers Analysis.

Evaluated performance metrics by comparing kids' product sales against overall brand performance.

Modeled future purchasing trends to estimate potential revenue loss, and inform inventory management decisions.

• Brooks Brothers' Women's Shoppers Analysis:

Segmented and analyzed female customer behavior by spend, age bracket, region, and channel (retail vs. digital).

Delivered insights on key performance indicators, such as repeat purchase rates, average spend per visit, and product performance, contributing to enhanced marketing strategies and revenue optimization.

• Endeavour Group Customer Analysis:

Performed multi-dimensional customer segmentation (including RFM and 9-box analyses) to identify revenue drivers and customer value segments.

Analyzed historical trends over three financial years to provide actionable insights on customer behavior, churn, and cross-brand purchasing patterns.

• Bobit:

Analyzed the customer data captured by a marketing management firm in Amperity to get actionable insights about customer engagement, survey response rate, and customer demographics.

Reported the same insights to the executive and management team of the firm, using Tableau and Databricks as the underlying architecture.

Involved in the end-to-end lifecycle of a Business Analyst project, from capturing and cleaning data to analyzing data and presenting the data to the end users of the insights.

Hyper-Personalized Marketing Tool.

(Streamlit, LLM, Python)

Built and deployed an LLM-powered marketing tool using Streamlit, with fine-tuned LLMs, processing over 100K customer records, reducing campaign creation time by 85%, and improving campaign performance by 15%.

MIDHANI

(Python, OpenCV, CNNs)

Developed an automated computer vision solution for analyzing metallographic images using Python and OpenCV, leveraging deep learning-based edge detection algorithms. Achieved a 99.8% reduction in image processing time, from 2 hours to 2 seconds.

• Worked in data preparation, computer vision model training in the cloud, and AWS management.
• Saved 28 labor hours per village with drone image analysis, by attaining 91% accuracy in object segmentation using deep learning in ArcGIS.
• Extracted and presented vital information on land encroachment, village facilities, development scores, and business analytics through a user-friendly dashboard for the Andhra Pradesh and Maharashtra governments.

Tools: Python, TensorFlow, Keras, AWS EC2, AWS S3, ArcGIS, Power BI.

• Implemented multimode ensemble machine learning and autonomous machine learning models for electrical protection systems.
• Worked on predictive maintenance using data science in the Battery Management System.
• Received a patent in the above-mentioned field, which was fully funded by ABB Global.

Tools: Python, Scikit-learn, XGBoost, Prophet