Shubham Kadam

Dynamic Senior Database Engineer with focus on creating scalable database solutions and optimizing workflows. Known for cross-functional collaboration, mentoring junior team members, and driving performance improvements that enhance system efficiency.

Schema & Trigger Management Framework — Python | Databricks | PostgreSQL | Slack

• Faced challenges with schema inconsistencies and trigger mismatches across multiple client databases.
• Built a Python-based Databricks framework to compare database schemas and validate triggers across 800+ databases.
• Automated detection of missing/modified triggers and schema changes, ensuring all client databases stayed in sync.
• Implemented Slack alerting system for real-time notifications on discrepancies and trigger failures.
• Result: Reduced manual validation time by 80%, ensured structural consistency, and maintained reliable database workflows.

Long Running Transaction (LRT) Analysis — New Relic | PostgreSQL  | Slack

• Faced recurring issues with idle and active long-running transactions, impacting performance.
• Developed a New Relic dashboard to identify and monitor LRTs in real time.
• Collaborated with PHP teams to resolve idle connections and close DB sessions properly.
• Analyzed active transactions, identified long-running queries, and recommended optimizations.
• Result: Improved system performance and reduced unnecessary DB locks.

Pg_stat_statement Performance Framework — Databricks | Python | PostgreSQL 

• Needed a way to detect queries causing performance delays in production.
• Built a Python-based framework in Databricks using pg_stat_statement data to identify queries running >5 seconds.
• Performed analysis, recommended optimizations, and sent proactive alerts to Slack channels.
• Result: Reduced query execution time from 5+ seconds to under 1 second, significantly improving system performance.

Automated Deployment Framework — Python| EC2 | PostgreSQL | Git | Slack

• Situation: Deployment of SQL scripts (DDL/DML/functions/views) across environments was manual, error-prone, and time-consuming.
• Task: Build an automated and reliable deployment system for SQL changes with review, validation, and alerting capabilities.
• Action:
  Designed two deployment frameworks — one for DDL/DML and another for functions & views.
  Implemented a review flow where SQL scripts were approved by the DB Engineering team before deployment.
  Automated deployment to lower environments and production using Git commands & batch processing.
  Set up Slack alerts for failure cases and deployment notifications.
• Result: Reduced manual deployment effort, ensured safer production releases, and enabled CI/CD-style database deployments with visibility and control.