Marcus T. Okafor

Profile

Senior SRE with 9 years scaling distributed systems at fintech and SaaS companies, specializing in Kubernetes orchestration, chaos engineering, and incident response automation. Reduced mean time to recovery by 61% at a 400-engineer org by rebuilding on-call runbooks and alert routing pipelines. Holds Google Cloud Professional DevOps Engineer certification and contributes actively to open-source observability tooling.

Work Experience

Senior Site Reliability EngineerMar 2021 - Present

Vantex Financial Technologies, Austin, TX

Redesigned Kubernetes-based deployment pipeline across 14 microservices, cutting release cycle time from 4 days to 6 hours and eliminating 3 categories of production rollback incidents annually.
Built a unified observability stack (Prometheus, Grafana, OpenTelemetry) that consolidated 7 disparate monitoring tools, reducing alert noise by 74% and saving the on-call rotation roughly 11 hours per engineer per month.
Led a cross-functional chaos engineering program using Gremlin, running 40+ fault-injection experiments per quarter that raised service resilience scores from 68% to 94% against SLA targets.

Site Reliability EngineerJun 2018 - Feb 2021

Loopline SaaS, Denver, CO

Automated incident triage with a PagerDuty and Runbook Automation integration, decreasing mean time to acknowledge (MTTA) from 18 minutes to under 4 minutes across a 24/7 production environment.
Migrated 60 TB of customer data from on-premises PostgreSQL clusters to Cloud Spanner with zero data loss and a maintenance window under 90 minutes, serving 2.1 million active users.
Defined and enforced SLO/SLA frameworks for 9 core APIs, resulting in 99.97% measured uptime over 18 consecutive months and a 22% reduction in customer-reported reliability tickets.

DevOps EngineerAug 2015 - May 2018

Brightpath Systems, Dallas, TX

Containerized a monolithic Rails application into 12 Docker services orchestrated with Nomad, reducing infrastructure costs by $340,000 annually through right-sized compute allocation.
Built CI/CD pipelines in Jenkins and later CircleCI for 6 product teams, reducing average deployment frequency from biweekly to daily and cutting failed deployment rate from 14% to 2.3%.
Instrumented application performance monitoring with Datadog APM, identifying 4 chronic latency bottlenecks that engineering resolved within one sprint, improving p99 response time by 48%.

Profile

Work Experience

Education

Skills

Links