I build distributed systems, AI infrastructure, and products.
I’m a Senior Member of Technical Staff at Oracle Cloud Infrastructure, where I work on AI infrastructure, distributed fleet health, and real-time visibility for GPU and bare-metal systems.
Currently, I’m building Pulse, a hardware health monitoring platform for 3M+ GPU accelerators and bare-metal servers across 100+ regions. I also build observability paths that process 5TB/day of telemetry for incident visibility.
Previously, I built financial-grade insurance systems at ACKO, including ledger systems handling 10M+ transactions per day, and worked on enterprise ordering, payment, and self-service systems at Amdocs.
June 2024 - PresentInitiated, defined, and led Pulse, OCI’s fleet health monitoring platform for 3M+ GPU accelerators and bare-metal servers across 100+ regions. Also led OCI observability work processing 5TB/day of telemetry, improving fleet incident investigation across 27 engineering teams.
August 2021 - June 2024Led the architecture and long-term evolution of Finacko, ACKO’s distributed financial ledger processing 10M+ transactions daily, reducing operational costs by 80%. Also built BigQuery configuration abstractions and a self-service Apache Airflow DLQ retry platform.
August 2019 - August 2021Owned development, deployment, and production reliability for payment and credit-check services in telecom ordering workflows. Improved data migration validation complexity by 40% and reduced chatbot API latency by 20% through compression and caching.
Co-founded and built a local e-commerce platform from scratch, leading backend architecture, product development, and engineering while raising seed funding from local vendors and growing the customer base by 20%.
Personal
In-progress AI operating system for India’s 900,000+ pharmacies, focused on prescription management, inventory, patient interactions, and compliance workflows.
Targets pharmacy workflows that still run on paper or legacy software, bringing AI assistance into day-to-day pharmacy operations.
OCI
Distributed hardware health platform monitoring 3M+ GPU accelerators and bare-metal servers across 100+ cloud regions.
Covers H200, GB200, GB300, and MI355X fleets, supports 107K+ OCI customers, and orchestrates hardware fault remediation workflows.
OCI
Real-time fleet observability platform for telemetry ingestion, query performance, and incident investigation at OCI scale.
Processes 5TB/day of telemetry, supports 27 engineering teams, and reduced MTTR from 8.31 to 5.24 hours.
ACKO
Finacko ledger platform for high-volume insurance finance, underwriting, and operations workflows.
Processes 10M+ transactions/day, eliminated manual finance operations, and reduced operational costs by 80%.
ACKO
Self-service Apache Airflow retry platform for failed event recovery beyond standard queue retention windows.
Eliminated AWS SQS’s 14-day retention limitation and let teams onboard queues through config changes.
Amdocs
Ordering, payment, credit-check, data migration, and chatbot API improvements for enterprise telecom systems.
Improved data migration validation complexity by 40% and reduced Multi-Channel Self Service API latency by 20%.