AJAY BONGANI

 Architecting enterprise AI platform infrastructure for Covasant — designing production MLOps pipelines using Kubeflow Pipelines, Argo Workflows, and MLflow for model training, experiment tracking, and registry management at scale.  Leading LLMOps and GenAI platform design — integrating LangChain, Azure OpenAI, and RAG-based retrieval architectures into CI/CD workflows with automated quality gates, observability, and governance guardrails.

 Delivered end-to-end Azure DevOps CI/CD architecture, AKS platform design, and SLO-based observability frameworks for UK fintech and e-commerce clients, achieving 40% faster deployment cycles.  Engineered ArgoCD GitOps delivery pipelines with automated rollback, blue-green gates, and burn-rate alerting — maintaining 99.97% uptime SLA and zero unplanned outages across 4 production environments.

 Architected and delivered a multi-cloud CI/CD platform (AWS + Azure) using multi-stage YAML pipelines, self-hosted agents, and GitHub Actions — enabling continuous AKS rolling deployments across 5 environments serving 200+ developers.  Built an Internal Developer Platform (IDP) with Bicep golden-path templates, reducing environment setup from 2 days to 2 hours and saving 1,000+ engineering hours per month while eliminating cross-team configuration drift.

 Architected Azure DevOps YAML pipelines with Bicep IaC for Oracle Cloud Kubernetes (OKE) and introduced Harness canary and blue-green progressive delivery with automated quality gates, reducing deployment incidents by 50%.  Designed and operated SLI/SLO monitoring using Prometheus and Grafana burn-rate alerting for 8 critical production services, achieving zero SLO breaches throughout the engagement.

 Engineered multi-environment Azure DevOps pipelines with Bicep and Terraform IaC, containerising enterprise workloads on AKS and reducing the release cycle from 2 weeks to 3 days.  Implemented Azure Monitor, Application Insights, and ELK Stack, along with automated Disaster Recovery runbooks, achieving Recovery Time Objective (RTO) of under 15 minutes.  Established Azure Arc hybrid governance across cloud and on-premises environments, cutting deployment failures by 60% across 3 production environments

 Led SRE operations across 50+ Linux servers (Rackspace) for 3 enterprise clients, implementing FinOps cost governance and self-healing incident runbooks — reducing downtime and infrastructure costs by 30%.  Designed Tekton GitOps canary pipelines with automated rollback triggers, significantly improving release reliability and reducing manual intervention in production deployments.

 Owned end-to-end SRE for a UK enterprise e-commerce platform on AWS, defining and operating SLOs with error budgets and integrating Dynatrace AIOps-driven incident response — reducing MTTR by 50%.  Managed EKS blue-green and canary deployments with automated rollback policies, achieving 50% faster release cycles and eliminating 70% of production rollbacks.

 Delivered GCP Cloud Build and Jenkins CI/CD pipelines, orchestrating GKE workloads via ArgoCD GitOps and achieving 30% container image size reduction.  Implemented Prometheus monitoring and Ansible automation, improving data pipeline reliability and operational visibility across GKE-hosted services.