I believe growth is a system, not a wish. My journey is about building habits the same way I
build infrastructure — with intention, monitoring, and continuous improvement. I treat
personal growth like infrastructure: built to last, monitored closely, and always iterating.
By day, I engineer cloud-native solutions at Novata. By discipline, I track my daily habits
and obsess over Kubernetes. My goal isn't just to write better code—it's to become a more
complete, effective, and intentional human being, every single day.
🔥
Consistency Over Perfection
Show up every day, even imperfectly
🧠
Systems Thinking
Build processes that make success inevitable
☁️
Cloud Native Mindset
Resilient, scalable, always monitored
📈
1% Better Every Day
Compound growth: small daily wins × time
FUN FACTS
🏀Lakers fan — Kobe #24 is the blueprint
🌆Based in Bangalore — Silicon Valley of India
☁️Targeting all 5 Kubernetes certifications (KCNA → CKS)
⏰Best
ideas before 9am — mornings are sacred
JOURNEY & TRACK RECORD
Experience & Education
My professional DevOps career timeline and academic achievements.
PROFESSIONAL EXPERIENCE
Feb 2022 - Aug 2025
DevOps Engineer
Busigence Technologies · Bengaluru
•Orchestrated Kubernetes clusters for container management, scaling, and maintenance, ensuring 99.9% uptime and efficient resource utilization.
•Designed and automated CI/CD pipelines using Jenkins, reducing deployment time by 40% and enhancing software delivery processes, while integrating GitHub for version control.
•Developed Docker images for diverse applications, optimizing resource usage and enhancing deployment consistency.
•Deployed containerized applications on Kubernetes, Docker, AWS, and Azure, leveraging cloud platforms for scalability and flexibility.
•Implemented cost optimization strategies on cloud platforms, resulting in a 30% reduction in infrastructure costs.
•Configured monitoring tools such as Prometheus, Grafana, and Loki for real-time tracking of application performance and infrastructure health, and incorporated YAML-based configuration for deployments.
Aug 2021 - Jan 2022
DevOps Engineer
Anuta Network · Bengaluru
•Administered Amazon EKS clusters, ensuring optimal performance and reliability with 99.9% uptime.
•Managed Kubernetes clusters on local environments and AWS, handling provisioning, scaling, monitoring, and troubleshooting tasks, reducing incident response time by 30%.
•Utilized automation tools including Jenkins, Docker, Kubernetes, Grafana, Jira, and GitLab, resulting in a 40% increase in deployment frequency and a 25% reduction in manual errors, while integrating GitHub Actions where applicable.
•Leveraged shell scripting to automate routine tasks, saving an average of 15 hours per week in manual effort.
•Ensured seamless integration of automation tools into DevOps practices, resulting in a 20% increase in team productivity and a 15% reduction in time-to-market.
•Conducted optimization and maintenance of EKS clusters, implementing best practices to enhance performance and reliability.
May 2021 - Aug 2021
DevOps Intern
Sirpi · Bengaluru
•Dockerized Node.js & Python applications.
•Deployed server-less projects on Cloud Run & Cloud Function using CI/CD.
•Worked on various cloud platforms including GCP, AWS, Digital-Ocean, and HETZNER.
•Managed new user access on Workspace by configuring permissions, roles, and security settings, streamlining the onboarding process and ensuring compliance with company policies.
ACADEMIC TIMELINE
2017 – 2021
Bachelor of Engineering (ECE)
Visvesvaraya Technological University (VTU) · Nitte Meenakshi Institute of Technology · GPA: 7.4
Systems, networks, signal processing, and engineering electromagnetics. Built a strong analytical foundation for complex engineering and cloud-native work.
ECEEngineeringVTU
🎯 KUBERNETES CERTIFICATION ROADMAP — CNCF
Targeting all 5 certifications in order of difficulty
⚡ 99.99% Mimir Uptime💰 -50% Storage Costs🚀 -95% Deploy Time
Mimir: Deployed
highly-available Mimir using Helm charts, integrating with Prometheus remote-write to
increase monitoring metrics accuracy by 20%. Secured dashboards with Grafana SSO (-40%
login time).
Tempo: Set up Tempo using the
Tempo Kubernetes Operator, slashing deployment provisioning times by 95%. Configured
dedicated storage buckets to cut tracing storage costs by 50%.
Loki: Deployed Loki with
Helm, configuring scalable ingress paths to reduce log query latency by 30%. Streamlined
log aggregation to speed up troubleshooting cycle times by 25%.