Headline

Data Engineer | Snowflake · dbt · AWS · Azure · Airflow · Data Vault 2.0 | Metadata-Driven Pipelines | PwC

About

I think a lot about how data actually moves through a system — not how it's supposed to move on paper, but what really happens when it hits production. For the last three years at PwC, that's been the job: building cloud data pipelines, migration frameworks, and data-quality systems across AWS, Azure, and Snowflake for banking and insurance clients. I came into data engineering from civil engineering. The habit that carried over: understand the whole system before you touch any single part of it. That's still how I approach a data platform. A few things I've built: → At Bank of Montreal — a metadata-driven L1→L2 migration framework on AWS Redshift. 25 target tables, 25 Airflow DAGs, and three dynamic stored procedures handling staging, DQ validation, and inserts. When the senior developers moved to other workstreams, I took over offshore delivery and finished ahead of the SIT/QA deadline. → At Great West Casualty — a QA framework for a 65-table Snowflake migration, built from scratch with no existing process. Parameterized SQL that scaled across every table from three inputs, and multi-level hash-comparison logic for the legacy validation workstream. I also use Generative AI in my day-to-day — specification parsing, automation, documentation — to compress repetitive work from days into hours. The focus is practical leverage, not AI for its own sake. The accountability for what ships stays with me. AWS Certified Cloud Practitioner Interested in scalable data platforms, workflow automation, and practical AI-assisted engineering. Open to Data Engineer and Data Platform roles at product companies in Bengaluru. Always happy to connect.

Location

Bengaluru

AI stack

cursor

opencode

claudecode

Work experience

Data Engineer — Farmers Insurance at PwC India, May 2026 - Present · 4 mos, Remote, Current engagement. Building the Silver-layer Data Vault 2.0 (Hub / Link / Satellite) on Snowflake with dbt for a P&C insurer's Business-Insurance policy domain — Workers' Compensation and Commercial Multi-Peril lines — within the team's metadata/registry-driven dbt framework. • Building Data Vault 2.0 models — 2 hub, 1 link, and 5 satellites so far — on Snowflake using dbt, within the team's metadata/registry-driven dbt framework • Applied core DV 2.0 patterns — business-key and grain analysis, hash-based surrogate keys, and incremental (merge) satellite loads with date standardization • Validated models with the framework's dbt tests — row-count parity and referential (hub ↔ satellite) integrity • Ramped onto dbt and Data Vault 2.0 hands-on on a live engagement where the methodology was new to the whole team

AWS Data Engineer — Bank of Montreal (BMO) at PwC India, Nov 2025 - Apr 2026 · 6 mos, On-site, BMO's Risk & Finance Data Hub runs on a multi-layer AWS Redshift architecture. PwC's scope was L1→L2 migration and L3 extract generation, working inside the client's Airflow, Glue DSF, and stored-procedure frameworks. • Designed and built a metadata-driven L1→L2 migration framework covering 25 target tables — table- and column-level metadata from DSIS specs drove source-to-target mapping, SCD2 handling, transformation logic, and incremental loads • Built 3 dynamic stored procedures (Staging, DQ, Insert), each orchestrated by a dedicated Airflow DAG — 25 DAGs total, with DQ-gated routing of passed/rejected records • Worked on the L3 extract workstream — config-driven pipelines using AWS Glue DSF (JSON Job + Job Step configs) across 4 extract groups and 30+ multi-GB files, with DQ validation, compression, and S3 encryption • Automated specification parsing with GenAI + Python — compressed roughly a week of setup into under a day • Took full offshore ownership after the senior developers transitioned, and delivered all L2 components ahead of the SIT/QA deadline — the BMO team sent a formal, team-wide appreciation note; I then moved to the L3 workstream

QA Engineer / Snowflake Developer — Great West Casualty Company at PwC India, Nov 2024 - Oct 2025 · 1 yr, Remote, On-premise-to-Snowflake migration for a US insurance carrier. No QA process in place and no experienced QA lead — I designed the framework end-to-end and owned it through delivery. • Built the QA methodology for a ~65-table migration from scratch — per-table test-case taxonomy, validation approach, and client-facing documentation standards • Engineered parameterized Snowflake SQL that ran on any table from just 3 inputs (database, schema, table) — eliminated manual query rewriting and compressed QA cycle time across the migration • Validated data systematically and surfaced 6–12 defects per table across the tables that carried differences (~35–45% of data) — collaborated with BA and Build teams on triage and resolution • Iterated QA documentation with GenAI — from developer-level output to client-readable reporting — earning senior-stakeholder recognition • Selected to independently design Legacy QA — multi-level hash-comparison logic (table → row → column) across DB2, Snowflake Clone, and Qlik extract sources

Education

Master of Technology - MTech, Construction Technology and Management, National Institute of Technology Karnataka

Bachelor of Technology - BTech, Civil Engineering, Tezpur University