Aashneil Uppuluri

Data-driven analytics professional with experience translating complex datasets into actionable insights that drive strategic decision-making and revenue growth. Proven track record of developing analytical models and tools to evaluate account performance, optimize pricing and contract strategies, and support multi-million dollar business outcomes. Skilled in collaborating with cross-functional stakeholders to align data insights with business objectives, while ensuring data accuracy, consistency, and scalability through robust validation and reporting processes. Proficient in Excel, SQL, and BI tools, with a strong ability to simplify complex information for both technical and non-technical audiences.

• Wine Quality Predictor, Utilizing a readily available dataset, Red Wine Quality, through Kaggle I was able to create predictor of wine quality with Python. Classification models were built in order to take into consideration many variables that would attribute to the quality being 'Great' or 'Poor'. The main objective of this project was to visually see the data and how it was being processed and interpreted. ML models were also used mainly to identify the errors between more/less node decision trees in terms of accuracy, and likewise, test sets and training sets were used to split data and compare against the models being used. Primarily relied on Python libraries like seaborn, matplotlib, and sklearn for visual plotting of data, and machine learning models.
• COVID-19 Analysis, The New York Times had an accessible data set that went in depth on the coronavirus counts in the United States at different levels. This complete record of the pandemic at the time made it a good example to properly analyze this data and create visual images to back up what I found. Breaking up specific variables such as number of positive tests, deaths, amount of results recorded, etc. and seeing their fluctuation over a given period of time with graphs made it possible to put these factors together to see correlations in specific areas of the United States. A key takeaway from this was that just because there was a downward trend of positive cases that didn't necessarily mean that this was an indication of an ending to the pandemic, more-so it meant that another wildcard would be thrown into the mix, such as state officials opening up access to public areas, which would then lead to a spike again, so this analysis was purely visual rather than predictive. Primarily relied on Python libraries like numpy, pandas, matplotlib.