Anmol Bajaj

I’m a Data Engineer with 1.8 years of experience building and operating cloud-native batch ETL pipelines on AWS. I specialize in extracting and transforming semi-structured data from MongoDB into analytics-ready Parquet datasets on Amazon S3 using AWS Glue and Python.

At Profectus Capital, I built and deployed a scheduled AWS Glue batch ETL pipeline processing 30K+ semi-structured MongoDB documents weekly into partitioned Parquet datasets. I engineered transformation logic for deeply nested JSON flattening, multi-collection enrichment, product ID-to-name mapping, and conditional schema normalization—then instrumented the pipeline with AWS CloudWatch to keep weekly runs stable across schema variations.

I’m proud that I reduced report generation time from 30–45 minutes to under 10 seconds by replacing manual extraction workflows with fully automated, scheduled transformations. I also built an internal Streamlit monitoring dashboard with role-based access control, and I’m expanding into data quality frameworks and streaming architectures while continuing to design production-grade data platforms.