My name is Vadim Pangaev, and I'm a Senior DevOps Engineer with a strong focus on building efficient delivery pipelines, scalable infrastructure, and AI-assisted automation across organizations.
I own infrastructure and delivery end to end: I set standards, make architecture decisions, and mentor engineers so teams can ship safely without waiting on me. I strive to deliver long-term improvements without sacrificing short-term agility. I take responsibility for my solutions and always aim to align infrastructure decisions with business objectives.
At ADIA Health, a fast-paced MedTech AI startup, I own infrastructure strategy and execution. I reworked the CI/CD pipelines using GitHub Actions and Terraform, reducing artifact size by 30% and dev iteration time by 50%. I designed and delivered a local development environment using Docker, deployable with a single command, to improve engineering autonomy and efficiency. I also built an event-streaming and CDC solution based on MSK Connect to increase cross-service data availability.
Prior to this, I worked as a Senior DevOps Engineer at EPAM Systems, where I supported diverse global clients across industries. My responsibilities included automating AWS infrastructure with Terraform and AWS CDK, optimizing deployment pipelines using Jenkins and GitHub Actions, and designing high-availability PostgreSQL clusters with hybrid-cloud replication strategies. These efforts led to improvements in scalability, availability, and delivery speed for both internal and production environments.
At Noveo Group, I focused on greenfield projects in the EU market, building DevOps practices from scratch to help clients reduce costs and reach their KPIs faster. I delivered Infrastructure as Code and CI/CD pipelines under strict budget constraints, helping small teams move quickly and safely.
Across EPAM and ADIA Health, I have also built AI-agent tooling for infrastructure work. At ADIA Health, Claude-based agents running on AWS triage pull requests, run Terraform plan and apply behind human approval in Slack, and answer developers' access and infrastructure support requests. They have handled 300+ PRs and 200+ requests. I use AI daily for IaC, CI/CD, and debugging, and keep it safe and affordable with shared agent rules, enforced permissions, and model tiering.

