At LIS Nepal, I designed and deployed ETL pipelines that processed 10M+ records daily using Python, Snowflake, Airflow, and Azure Data Factory. I also validated 100M+ records for a Netezza-to-Snowflake migration.
I closed a daily sales reporting gap with hourly Snowflake Tasks and stored procedures, and automated file preprocessing and secure SFTP delivery in Airflow, cutting each run from 30 minutes to under one minute.
At the University of Wyoming, I built a ruler-calibrated pipeline that reconstructs strawberry plants from ordinary camera video and segments their leaves, achieving under 5% height measurement error. I also designed Point-MAE training objectives that improved generated-point coverage on Pepper-4D from 9.5% to 20.3%.
My work spans data engineering and 3D computer vision. I hold AWS Data Engineer – Associate and AWS AI Practitioner certifications, and my research includes publications on plant point-cloud reconstruction and 3D referring expression segmentation.

