Analytics Engineer · pipelines, warehouses and streaming · Dhaka, Bangladesh

Nearly five years building the data infrastructure behind analytics, currently as a Senior Data Analyst (Analytics Engineering) at Optimizely. My day-to-day is production engineering: Python ingestion into Snowflake on Kubernetes, tested dbt marts, cost-controlled warehouses with Terraform, and CI that builds and diffs changes against production before review.

Right now: building an end-to-end streaming lakehouse for telecom usage records: Kafka into Spark Structured Streaming and Iceberg on HDFS/YARN, with idempotent writes, deduplication, late-arriving data handling and safe day-level backfills.

Daily use

Building now

Also used

• Quote-to-revenue reconciliation: sole developer of a quarterly dbt build (17 models, 13 matching macros, primary-key tests across 20+ source tables) that reconciles multi-year quotes to signed-contract terms, raising the match rate from 61% to 92%.
• Finance close: built the tested dbt suite behind the monthly close, cutting the manual rebuild from 4–5 days to about 1, and rewrote a 14M-row shared model with zero differences across 9.6M rows compared.
• Warehouse cost control: designed the analyst Snowflake warehouse and its cost monitor in Terraform, shipped through GitHub Actions: auto-suspend, query timeout and a daily credit cap with alerts.
• Safer changes: Python pre-push hooks that build each changed dbt model against production and diff row counts and nulls before review.