Provisioning and lifecycle management of cloud infrastructure using AWS CloudFormation, Terraform, and AWS CDK — ensuring consistency and scalability across environments.

Designing Disaster Recovery and Automated Failover mechanisms for traffic evacuation during outages.

Ensuring system resiliency and minimal downtime to meet Clever's SLO of 99.98% and supporting more than 27 million requests per month.

Migrating Clever's observability platform from Datadog to Clickhouse that increased log availability

duration by 20x while keeping logging costs the same.

Building infrastructure tooling that streamlines the entire service lifecycle, from local development to production deployment. Reducing friction for engineers shipping code across 400+ microservices.

Performance engineering efforts — including targeted load testing and tuning — which improve system uptime and reduced operational costs at scale.

Working on a next-gen Opensearch product. Building infrastructure using distributed systems, cellular architecture, designing routing mechanisms, gateways and AWS tools. Wear multiple hats with debugging, ops and peer reviews as part of the role.

Own and develop the Advanced Threat Defense(ATD) sandbox licensing system which is used to generate and renew licenses for both physical and virtual ATD devices catering to 2000+ customers. Uses Celery for distributed task queuing with RabbitMQ messaging queue(Python, Django framework, MySQL, RabbitMQ)

Worked on implementation of Elasticsearch indices that worked across multi node ATD sandbox clusters for data synchronization and malware-based searching. The framework could support upto 8 nodes with data of 3 million samples each. (Elasticsearch)

Developed a Threat Timeline feature in Advanced Threat Defense (ATD) sandbox reports that tracks periodic malware events, up to 1000+ for each sample, and represents them visually. (Python, HTML/CSS, Highcharts)

Implemented features in the Summary module for ATD including redesigning Analysis Reports and mapped traditional ATD malware behaviors into 200+ MITRE ATT&CK Techniques. The rollout of this feature made ATD one of the leading malware products to use the emerging MITRE framework (Python, C++)

Built an automation engine which detected and fixed vulnerable JEXL expressions with 70% efficiency

Worked on product features like Currency and Application Transactions for new version roll-out

Designed a module to track JUnit testing progress for incremental ServiceNow platforms which exposed unit-test coverage levels and areas for improvement

Collaborated on Advanced Threat Defense- Family Classification; the module’s algorithm improved advanced malware detection for ATD

Designed a Django-based web module that served malware information and post-analysis database for 2 million+ entries

Implemented Remediation Kit functionality which measures the effect of malicious files on consumer devices and attempts to reverse the damage caused by 40+ families of malware