At TechOptima, I worked on an AWS data quality and metadata project using Amazon Bedrock to generate column descriptions and plain-language summaries of data issues. I also built and maintained Glue and PySpark jobs that prepared data for Amazon S3 and Amazon Redshift.
I used Lambda, EventBridge, and Step Functions to trigger processing when data arrived, and added checks for nulls, duplicates, and schema changes. I wrote prompts and JSON formats for consistent LLM responses, then reviewed generated summaries with analysts before use.
Earlier, I developed cloud ETL pipelines and streaming workflows at Norstella and CAI, and supported data extraction, reporting, and pipeline work at Glowbrite Systems. My experience includes AWS data services, Snowflake, Apache Spark, and preparing datasets for analytics and AI use cases.
