Sep 2019 – Present · 6 yrs 6 mos

• Led Linkedin's search infrastructure, supporting 100+ verticals with traffic
• of 10,000 QPS, maintaining 99.99% uptime.
• Automated OS upgrades for 50,000+ servers, later scaling the framework
• to 200,000 servers, reducing downtime and operational workload.
• Collaborated with Dell to resolve DIMM issues, resulting in $1M cost
• savings across 5,000 hosts.
• Optimized search performance through Java profiling, JVM tuning, and
• load testing, significantly reducing hardware costs.
• Built CLI tools for compliance tracking, delivering real-time insights for
• faster issue resolution.
• Designed a Go-based metric ingestion system to achieve low-latency
• performance and enhance observability.
• Managed large-scale deployments and CI/CD pipelines, improving release
• efficiency and SLA compliance.
• Spearheaded incident response and postmortem analyses, driving
• process improvements and reducing recurring issues.
• Key Achievement:
• Reduced operational cost and improved search performance by 50%
• through system optimization.

Nov 2015 – Sep 2019 · 3 yrs 10 mos

• Managed large-scale data ingestion pipelines, integrating diverse sources
• like Kafka, Oracle, MySQL, and third-party vendors.
• Designed and implemented the China Data Analytics pipeline for seamless
• multi-colo Hadoop operations.
• Built tools for efficient debugging and implemented proactive alerting
• and monitoring systems.
• Ensured data reliability for executive-level reporting, meeting strict SLAs.
• Key Achievement:
• Developed tools that improved triage efficiency by 50%, for dataset
• availability for China Data Science and Machine Learning Team for
• analytics
• Reduced SLAs from 2 weeks to 2 days for new dataset onboarding for
• China