Jan 2025 – Present · 1 yr 2 mos · Remote

• Execution-focused leadership within the broader SRE organisation. Manage deployment scheduling, implement core SRE principles including SLOs/SLIs, and lead multiple SRE-driven initiatives to reduce COGS and improve infrastructure efficiency.
• Accomplishments:
• Migrated workloads on Azure and GCP to AMD-based instances, enabling SMT and mitigating processor side-channel vulnerabilities: This led to faster HTTP response times, lower CPU usage across regions, and the decommissioning of 20–40% of hosts—achieving significant performance gains and cost savings.
• Structured and enriched monthly SRE updates to executives—improving visibility into infrastructure health, incident trends, reactive vs. proactive workload balance, and key team operational metrics.
• Defined and implemented SLOs/SLIs across the edge, orchestration, and storage layers—establishing clear reliability targets and driving a culture of accountability.
• Infrastructure scale:
• → Adobe commerce cloud (Platform.sh managed): 25,000+ VMs across public cloud providers (AWS, Azure, GCP).
• → Adobe starter commerce cloud: Kubernetes-like PaaS platform orchestrating 1,800+ VMs and 200,000+ LXC containers.

Jan 2022 – Dec 2024 · 2 yrs 11 mos · Remote

• Led a 35+ member team of engineers and SREs—including 3 direct-reporting managers—focused on improving platform stability and reducing COGS across Platform.sh’s PaaS container infrastructure.
• Accomplishments:
• Co-developed with our CTO a novel TLS-layer concurrency limiter, the first of its kind, which controls concurrent TLS handshakes using the SNI header. This significantly improved the platform's ability to mitigate abuse and DDoS attacks—reducing monthly incidents from 1–2 to virtually zero. One of my most impactful and rewarding contributions, with lasting effect on platform resilience.
• Resolved a Ceph RBD volume leak caused by Copy-on-write (CoW) snapshot relationships, preventing $9,000+/month in storage waste.
• Conducted a thorough review and recommended removal of two disaster recovery snapshots without raising Ceph replica counts, leveraging existing data replication across AWS partition placement groups. This optimization delivered over $1M in cost savings improving the company margins by 2-3 basis points!
• Established a regular release cadence for platform components, improved testing and validation processes, and introduced phased rollouts—accelerating delivery velocity while maintaining platform safety and reliability.

Jul 2018 – Dec 2021 · 3 yrs 5 mos · Remote

• Led a team of 12 high-performing SREs across Asia Pacific and EMEA, building and operating a modern, hybrid-cloud, container-based PaaS.
• Accomplishments:
• Grew the APAC SRE team from 2 to 6 engineers, and within a year, took on leadership of the EMEA region — managing a combined team of 12 SREs across multiple time zones.
• Designed and built the first PaaS region upgrade automation, reducing upgrade time by 50% (from 8 hours to 4) by automating a complex 20-step manual workflow.
• Led the rollout of enterprise and standard stacks on Google Cloud and OpenStack, modernizing platform support and tooling to enable the company’s next wave of growth across multiple cloud environments.
• Key role in shaping platform direction — contributing to policy decisions around capacity planning and cost optimization to help manage and reduce COGS.

May 2016 – Jun 2018 · 2 yrs 1 mo · Remote

• First SRE hire. Heavy hands-on. Fast-moving. Zero red tape. Built core infra that scaled with the company.
• Accomplishments:
• Built the first automation tools for backups and instance resizes — taking repetitive, manual work off engineers' plates from day one.
• Developed Python-based enterprise provisioning tools for AWS (EC2, EBS, VPC, S3), cutting environment setup time from 5 hours to just 45 minutes. We went from provisioning 1–2 environments per day to several — massively accelerating provisioning velocity.
• Enterprise stack bring up on Azure using Virtual Machines, Azure blob storage, ARM templates, and un-managed disks — unlocking million dollar enterprise deals.
• Handled gnarly L3 support escalations, staying close to production and solving the most difficult customer issues.