Prachi M

Senior SRE and Platform Engineer with 12+ years of experience designing, scaling, and operating cloud-native infrastructure across multi-cloud environments (AWS, GCP, Azure, OCI). Deep expertise in Kubernetes cluster lifecycle management, CI/CD automation, observability, and infrastructure as code. Proven track record of managing 56+ production Kubernetes clusters with >99.95% availability and reducing cloud spend by up to 35%.

Dual Kubernetes certified (CKA, CKS) and Terraform certified. Experienced in SLI/SLO ownership, on-call incident response, blameless postmortems, and mentoring engineers in platform best practices.

Senior Site Reliability Engineer at AVALARA (2025-02 – Present)

• Owned end-to-end platform reliability for a fleet of 56+ Kubernetes clusters across AWS (EKS), GCP (GKE), Azure (AKS), and OCI spanning 6 environment tiers, with SLO targets anchored at >99.95% availability.
• Diagnosed and resolved a systemic DNS resolution bottleneck affecting microservice communication at scale - designed and rolled out a NodeLocal DNSCache architecture for CoreDNS that eliminated timeout-class errors and became the standard for all new cluster provisioning.
• Established SLI/SLO frameworks for foundational cluster services (CoreDNS, NodeLocalDNS); owned error budget tracking, led structured incident response during on-call, and drove RCAs to permanent infrastructure-level fixes rather than workarounds.
• Architected cluster-wide observability using Prometheus, Grafana, Alertmanager, and OpenTelemetry - moving the team from reactive alert-driven response to proactive anomaly detection, reducing MTTD by ~50%.
• Led end-to-end migration from NGINX Ingress to Envoy Gateway, modernising the traffic layer with L7 routing policies, mTLS enforcement, and improved observability at the ingress level.
• Deployed a Kubernetes MCP server across the cluster fleet, enabling context-aware querying of incidents and giving engineering teams a self-service interface for faster cross-cluster troubleshooting without SRE dependency.
• Drove cloud cost optimisation using Karpenter for intelligent node provisioning and right-sizing across multi-cloud clusters, achieving 20-35% reduction in cloud spend while maintaining performance SLAs.
• Standardised cluster lifecycle operations by authoring upgrade runbooks, rollback playbooks, and change management gates across all 56+ clusters, reducing unplanned downtime during upgrades by ~40%.
• Mentored a junior SRE through the full Envoy Gateway migration, owning the technical design and using the project as a structured learning path to build the engineer's independent platform capability.

Senior Software Engineer at VISA (2021-11 – 2025-02)

• Overhauled the team's CI/CD practice in Jenkins - replaced ad-hoc freestyle jobs with versioned shared pipeline libraries and pipeline-as-code, creating a consistent, auditable deployment standard across payment processing services and achieving ~2x faster deployment cycles.
• Led containerisation of payment applications on CloudView, decoupling deployments from environment-specific configuration and improving release velocity by ~40% while eliminating a class of environment-drift failures that previously delayed releases.
• Designed a Test Automation Framework in Python and Robot Framework embedded directly into CI pipelines - increased test coverage by ~50%, cut manual QA effort by ~70%, and shifted quality validation left so defects were caught before staging.
• Collaborated with business and product teams to translate payment workflow requirements into automated pipeline quality gates, ensuring production deployments met compliance and correctness standards for critical transaction flows.
• Mentored 2 engineers in automation engineering and CI/CD ownership, enabling them to independently build and extend the framework - reducing onboarding time for new automation tasks by ~60%.

Automation Engineer at STANDARD CHARTERED BANK (2018-08 – 2021-11)

• Drove the bank's shift to infrastructure as code using Terraform, replacing manual server provisioning with version-controlled, peer-reviewed infrastructure definitions - cutting environment promotion time from development to QA by ~90% and introducing a change audit trail required for regulatory compliance.
• Designed and owned CI/CD pipeline architecture that enforced standardised build, test, and deployment stages across teams - replacing inconsistent manual release processes with governed, repeatable workflows that reduced release cycle time significantly.
• Authored Terraform module governance standards and a shared module library used across dev, QA, and staging environments, eliminating configuration drift and cutting drift-related production incidents by ~45%.
• Collaborated with business stakeholders to translate operational and compliance requirements into infrastructure and pipeline design decisions, ensuring delivery against defined SLA thresholds.
• Embedded automated regression testing into release pipelines, providing a quality gate that caught environment-specific defects early and markedly reduced production bug rates across banking services.
• Mentored team members on IaC principles, pipeline design, and automation best practices - building team self-sufficiency and reducing reliance on manual, knowledge-siloed deployment processes.

Associate Consultant at CAPGEMINI (2014-03 – 2018-08)

• Built the automation infrastructure foundation for multiple client engagements using Docker, containerising test environments to eliminate the "works on my machine" problem and cutting application build times by ~35%.
• Wrote JMeter performance test suites to benchmark API throughput and latency under load, surfacing bottlenecks before production and improving system resilience across client deliverables.
• Automated UI regression testing using Selenium and TestNG, replacing manual test cycles with a repeatable suite that reduced testing overhead by ~50% and gave development teams faster feedback on regressions.
• Integrated testing tooling, build systems, and deployment scripts into a unified automation framework, creating a single pipeline entry point for build, test, and package stages across multiple concurrent projects.

Bachelor of Engineering in Electronics & Communications – Jain University (2013)