Snowflake platforms that run fast, stay locked down, and cost less to operate.

Three queries I run first on a Snowflake account

The Account Usage views show most of what's wrong with a platform. Pick a query to see what it surfaces and what I do next.

-- Weekly credits per warehouse, last three months
SELECT warehouse_name,
       DATE_TRUNC('week', start_time) AS week,
       SUM(credits_used)              AS credits
FROM   snowflake.account_usage.warehouse_metering_history
WHERE  start_time >= DATEADD('month', -3, CURRENT_TIMESTAMP())
GROUP BY 1, 2
ORDER BY credits DESC;

-- Queries that spilled to remote storage this week
SELECT query_id,
       warehouse_name,
       warehouse_size,
       bytes_spilled_to_remote_storage,
       total_elapsed_time / 1000 AS seconds
FROM   snowflake.account_usage.query_history
WHERE  bytes_spilled_to_remote_storage > 0
  AND  start_time >= DATEADD('day', -7, CURRENT_TIMESTAMP())
ORDER BY bytes_spilled_to_remote_storage DESC
LIMIT  20;

-- Everyone who currently holds ACCOUNTADMIN
SELECT grantee_name AS user_name,
       granted_by
FROM   snowflake.account_usage.grants_to_users
WHERE  role = 'ACCOUNTADMIN'
  AND  deleted_on IS NULL
ORDER BY user_name;

What I work on

Performance tuning

Query History, Query Profiles, and Account Usage analysis. Micro-partition pruning, clustering, caching, spilling, queuing, and concurrency. Multi-cluster warehouses and Query Acceleration Service.

Cost governance

Resource Monitors, budgets, alerts, and query-level cost attribution. Showback and chargeback reporting, with serverless and Cortex AI consumption included.

Security and governance

Least-privilege RBAC and database roles. Dynamic masking, row access policies, network policies, SSO and SCIM, access auditing, and AI guardrails for governed AI workloads.

Pipelines

Snowpipe, Streams, Tasks, Dynamic Tables, and dbt for incremental ELT and CDC. Schema evolution, data validation, and ingestion monitoring.

Cloud and automation

AWS (S3, IAM, KMS, VPC, PrivateLink, RDS, DMS, CloudWatch), Terraform, CloudFormation, Python, shell, Snowflake CLI, Git, and CI/CD with change controls and rollback.

Reliability and databases

Incident management, root cause analysis, runbooks, HA and DR, failover groups, Time Travel, and zero-copy cloning. Oracle RAC, GoldenGate, Data Guard, PostgreSQL, and SQL Server.

Experience

1. Sep 2025 to present

   Senior Cloud Database Administrator / Engineer, VNS Health

   • Cut credit spend by 40% by right-sizing multi-cluster warehouses and isolating workloads, and cut query runtime by 50% by tuning from Query Profiles.
   • Lead cost governance with Resource Monitors, budgets, alerts, and showback and chargeback reporting, including serverless and Cortex AI consumption.
   • Implement least-privilege RBAC, masking and row access policies, SSO and SCIM, and access auditing. Provision with Terraform and Git-based CI/CD.
   • Build observability dashboards and alerting for warehouse utilization, ingestion latency, task failures, and data freshness. Led onboarding of the Espresso optimization platform.

   Stack: Snowflake, Cortex AI, AWS, dbt, Python, Terraform, Snowflake CLI, PostgreSQL, SQL Server.

2. Jan 2025 to Sep 2025

   Senior Cloud Database Administrator / Engineer, Capital One

   • Administered Snowflake, PostgreSQL, SQL Server, and Oracle on AWS, including warehouse configuration, tuning, RBAC, and compute cost optimization.
   • Led on-premises to AWS migrations with DMS, Schema Conversion Tool, and CDC, including validation, cutover planning, and recovery procedures.
   • Built highly available RDS with multi-AZ, backups, encryption, and DR controls. Diagnosed PostgreSQL regressions with EXPLAIN ANALYZE.

   Stack: Snowflake, AWS, RDS, DMS, PostgreSQL, SQL Server, Oracle, GoldenGate, Terraform, CloudFormation, Python.

3. Jan 2023 to Dec 2024

   Senior Database Administrator / Engineer, Verizon

   • Optimized Snowflake warehouse utilization and SQL performance using Query History and Query Profiles. Managed user provisioning, RBAC, and grants.
   • Engineered Oracle RAC, GoldenGate, and Data Guard for high availability and disaster recovery on mission-critical applications.
   • Automated monitoring, incident diagnostics, and maintenance with SQL, PL/SQL, Python, and shell.

   Stack: Snowflake, Oracle, RAC, GoldenGate, Data Guard, SQL Server, AWS, PostgreSQL, Python.

4. Apr 2018 to Dec 2022

   Database Administrator / Engineer, Fannie Mae

   • Administered enterprise Oracle across production, development, and test, with RAC, Data Guard, RMAN, patching, and DR readiness.
   • Ran GoldenGate replication for real-time sync, migrations, and Oracle-to-PostgreSQL integration.
   • Supported on-premises to AWS migrations with DMS and RDS, including schema assessment and data validation.

   Stack: Oracle, RAC, Data Guard, GoldenGate, RMAN, AWS, DMS, RDS, PostgreSQL, Python.

5. Dec 2016 to Mar 2018

   Database Administrator / Engineer, Virani

   • Administered Oracle databases, including backup and recovery, access controls, and monitoring with Oracle Enterprise Manager.
   • Developed and tuned SQL and PL/SQL, and automated health checks with Python and shell.

   Stack: Oracle, SQL, PL/SQL, Python, Linux, Oracle Enterprise Manager.

Selected work

Snowflake cost optimization

Cut credit spend by 40% through warehouse right-sizing, tighter auto-suspend, workload isolation, and Resource Monitors. Added query-level attribution for team showback and chargeback.

Query performance tuning

Cut query runtime by 50% by reading Query Profiles and fixing pruning, clustering, spilling, and queuing.

Snowflake observability

Dashboards and alerting for warehouse utilization, query performance, ingestion latency, task failures, and data freshness.

Oracle to AWS migration

Designed and ran Oracle-to-AWS migrations with DMS and GoldenGate, including validation, cutover planning, and recovery procedures.

Predictive monitoring

Python-based anomaly detection that flags performance problems before they reach production.

Side projects

A job aggregation tool that writes daily listings to Excel, and PickStrike.com, a PostgreSQL-backed web app for IPL match predictions.

Credentials