Berkay Çelik

Managed SRE operations for a high-volume travel platform, leading cloud migration efforts and architecting MLOps solutions.

• Managed SRE operations for a travel platform processing over 1.5B annual transactions, achieving 99.99% uptime through Prometheus/Grafana monitoring and automated incident response.
• Led a critical Azure cloud migration for 200+ microservices on OpenShift, reducing infrastructure costs by 30% and deployment time by 60% using Terraform IaC and CI/CD pipelines.
• Implemented a comprehensive observability framework with 50+ custom Prometheus metrics and Grafana dashboards, resulting in a 40% decrease in Mean Time To Resolution (MTTR).
• Architected an MLOps platform utilizing Kubeflow and MLflow for AI-powered recommendations, streamlining model release cycles from weeks to just 2 days.
• Collaborated cross-functionally with 25+ engineers across 4 time zones to establish robust SRE best practices, achieving 95% error budget compliance.
• Automated infrastructure provisioning using Azure ARM templates and Ansible, reducing manual tasks by 70% and enhancing operational efficiency.

Orchestrated cloud deployments for a supply chain platform, automating infrastructure and modernizing CI/CD pipelines.

• Orchestrated complex cloud deployments for a supply chain platform serving 50+ enterprise clients, managing Azure infrastructure and reducing associated costs by 25%.
• Automated infrastructure provisioning using Terraform and Ansible, drastically cutting deployment time from 4 hours to 30 minutes.
• Designed and implemented a robust RBAC framework with 20+ custom roles, reducing access provisioning time by 75% and ensuring SOC2 compliance.
• Implemented a comprehensive monitoring stack with Prometheus/Grafana, improving system availability from 97% to 99.5% and reducing MTTR by 40%.
• Modernized CI/CD pipelines using Azure DevOps, implementing blue-green deployments to enable zero-downtime releases and enhance system stability.

Optimized performance for enterprise applications and established SRE practices.

• Optimized performance for 20+ enterprise applications using Instana APM, reducing P1 incident resolution time from 2 hours to 45 minutes.
• Deployed and managed Kubernetes clusters on RedHat OpenShift with auto-scaling and self-healing capabilities, increasing system uptime to 99.5%.
• Established and enforced SRE practices, including golden signals monitoring and error budgets, which led to a 40% reduction in Mean Time To Resolution (MTTR).

Migrated applications to AWS, developed Terraform modules, and optimized serverless architecture.

• Successfully migrated over 15 applications from on-premises to AWS (EC2, RDS, S3), resulting in a 30% reduction in monthly costs, saving $8K per month.
• Developed comprehensive Terraform modules for AWS infrastructure, automating the deployment of over 50 resources and accelerating development cycles.
• Implemented robust CI/CD pipelines using Jenkins and AWS CodeDeploy, achieving a 95% deployment success rate for critical applications.
• Optimized serverless architecture leveraging AWS Lambda and API Gateway, significantly improving application response times from 800ms to 200ms.

Delivered lab sessions, developed automated grading systems, and mentored teaching assistants.

• Delivered over 20 lab sessions focused on optimization techniques to 150+ students, achieving a 92% student satisfaction rating.
• Developed an automated grading system using Python and VBA, which reduced grading time by 60% and improved efficiency.
• Mentored and trained 15+ teaching assistants through structured programs, enhancing their instructional capabilities and team performance.

Engineered deep learning models (Informer, Autoformer) to predict BTC/ETH prices using Twitter sentiment analysis.

Built a decentralized e-commerce platform using React, Node.js, and Solidity smart contracts.