Nikhar Gaurav

• Designed and operated large-scale Spark (Scala) data pipelines processing high-volume, mission-critical datasets with strict SLA requirements.
• Built and maintained a scalable ETL platform supporting multiple internal consumers across finance and analytics domains.
• Migrated legacy Oracle and Redshift workloads to an AWS-based data lake architecture, improving scalability, reliability, and operational efficiency.
• Developed and deployed Airflow DAGs with custom operators to dynamically fetch configurations from DynamoDB, enabling centralized control and faster deployments.
• Implemented data availability, validation, and upstream dependency checks using AWS Lambda to proactively detect data issues.
• Improved pipeline performance by ~50% by analyzing Spark UI metrics, resolving data skew, memory pressure, and inefficient joins.
• Designed a GUID-based processing framework to replace timestamp-based tracking, improving idempotency and failure recovery.
• Established monitoring, alerting, and ticketing workflows using AWS CloudWatch to reduce incident response time.

Operational Ownership, User Support & Business-Critical Processes

• Led a 7-member support team (4 contractors, 3 support engineers) supporting ~1,200 internal users.
• Owned and executed business-critical operational processes including: Month-end close, Data restatements, Snapshotting and historical backfills and Planning and forecast data refreshes
• Ensured data accuracy, completeness, and timeliness for close and planning cycles with zero missed deadlines.
• Owned ticket triage, prioritization, escalation, and resolution for data access, query optimization, and pipeline issues.
• Reduced average ticket response time from ~6 hours to through improved triage and operational processes.
• Reduced average ticket resolution time from ~32 hours to , significantly improving SLA compliance and user satisfaction.
• Closed ~650 tickets in the past year and conducted 120+ office hours to proactively address recurring issues.
• Held daily standups to review backlog, assign work, and remove blockers.
• Delivered bug fixes and small feature enhancements in parallel with core data engineering development.
• Served as primary incident point of contact during high-severity issues, coordinating resolution and communicating status to stakeholders.

• Built and maintained cloud-scale telemetry and analytics pipelines using Azure Data Explorer (Kusto) and KQL.
• Queried and analyzed high-volume datasets stored in Azure Cosmos DB to generate near real-time operational insights.
• Developed ingestion and transformation workflows using Azure Data Factory and Azure Databricks.
• Built analytics pipelines to surface operational risk signals and underperforming customer segments.
• Implemented ML models (XGBoost, LightGBM) that improved Azure customer satisfaction score by 0.2.
• Applied SHAP-based explainability to identify drivers of customer dissatisfaction and guide remediation efforts.
• Created dashboards using Azure Data Explorer and Power BI for engineering and operations stakeholders.
• Partnered with engineering and support teams to translate insights into platform and process improvements.

• Built end-to-end portfolio analytics and back-testing pipelines using R and SQL.
• Converted legacy MATLAB models to automated R-based workflows.
• Developed R Shiny dashboards for fixed-income investment KPIs.
• Implemented batch processing pipelines for financial datasets.

• Developed and supported enterprise data warehouses for a global Unilever platform spanning 16 countries.
• Built ETL pipelines and optimized SQL/SSIS workflows, reducing system response time by 50%.
• Performed recurring data quality validation and KPI reconciliation.
• Received multiple awards for large-scale international Data warehouse implementations.
• Created on-demand ad-hoc reports for KPI reporting and generating data insights by comparing data using SQL, SSRS and SSAS cube.
• Performed data migration for OLTP and OLAP system through ETL package.
• Performed monthly data testing and quality check on databases within data warehouses to reduce data redundancy and remove faulty data impacting various business KPIs.