Krishna A

About Krishna A

Most teams don’t fail at building software.They fail at scaling it.I work on the part that breaks when things grow.Over the last 7+ years, I’ve built and scaled infrastructure across AI startups, semiconductor HPC, and enterprise SaaS.At Web AI, I own infrastructure, CI/CD, and observability across multiple product lines powering AI workloads in production.Before that at NXP Semiconductors:• Managed 500+ node Kubernetes clusters for chip design and HPC simulation• Migrated 300+ applications to AWS, reducing infrastructure cost by 35%I started my journey at NCR as a sysadmin, where I learned how fragile systems can be and how important automation is at scale:• Automated patching across servers• Built CI/CD pipelines that reduced release cycles from weeks to daysAlong the way, I’ve focused on one thing: Building systems that scale reliably under real-world pressureThat means:• Strong DevOps and platform engineering foundations• Secure, scalable multi-cloud architecture• High-visibility observability and faster incident response• Infrastructure that supports real business growth, not just deploymentsKey outcomes:• 30–35% reduction in cloud costs• 40% faster CI/CD pipelines• 99.95% uptime for production systemsTools I work with:Terraform, Kubernetes, AWS, Ansible, Helm, ArgoCD, GitLab CI/CD, Jenkins, Vault, Prometheus, Grafana, Datadog, DockerCertified Kubernetes Administrator (CKA)AWS DevOps Engineer – ProfessionalM.S. Computer Science, Pace UniversityIf you\'re building systems that need to scale without breaking, we should talk. k••••••••@gmail.com

Experience

1. Senior Devops Platform Engineer

   webAI

   Sep 2025 — Present · Austin, TX, US

   Owned end-to-end infrastructure, CI/CD, and observability for a $2.5B AI startup across training, inference, and platform systems.• Built and managed infrastructure across dev, staging, and production using Terraform, Kubernetes, and Helm• Reduced cloud spend by 30%(~$18K/month) through cost governance and resource optimization• Improved CI/CD performance by 40% across 40+ repositories, reducing pipeline runtime from 35 → 21 minutes• Cut environment provisioning time from 4 hours → under 20 minutes using reusable Terraform modules• Enabled secure enterprise connectivity using AWS PrivateLink, VPC peering, and VPNs (no public data exposure)• Maintained 99.95% uptime with centralized observability (Grafana, Datadog, CloudWatch)• Reduced model build failures by 65% through pipeline reliability improvements

Education

• Anna University Chennai

  Bachelor of Engineering - BE, Computer Science

• Pace University - Seidenberg School of Computer Science and Information Systems

  Master of Science - MS, Computer Science