Abhishek Pundir

Spearheading the development of ELT pipelines, effectively acquiring SQL Server data and optimizing storage in Snowflake, leveraging expertise in data warehouse design. Collaborating with multi-disciplinary teams to understand business requirements. Analyzing statistical data, cleaning & preprocessing, and visualizing data. Identifying and implementing statistical models including linear models, multivariate analysis, stochastic models, sampling, optimization, and time series analysis. Fostering collaboration to develop and deploy impactful machine learning solutions using Python, covering the entire end-to-end ML cycle. Streamlining the end-to-end ML solution workflow through automation using Apache Airflow. Conducting statistical analysis and data mining to gain insights into market behavior and consumer trends. Achievements include improving claim outcome prediction by 82%, deploying and maintaining real-time Machine Learning models on AWS Cloud, and successfully implementing a Python-based time-series model (fbprophet) to forecast non-alcoholic beverage consumption.

Spearheading the development and execution of Python-based ETL pipelines to extract data from diverse sources (emails, S3 bucket, Google Analytics, social media). Streamlining data extraction through a CRON job. Conducting comprehensive data preprocessing and analysis. Actively partnering with cross-functional teams to strategize and implement A/B tests, resulting in a 20% enhancement in conversion rates. Developed and implemented Python-based ETL pipelines loading data into SQL Server. Achievements include deploying and maintaining a real-time machine learning model (XGBoost) on AWS Cloud, and developing, optimizing, and assessing ML models (XGBoost, LightGBM, and Random Forests) to enhance claim cost accuracy by 75%.