DevOps Intern (AI Infrastructure)

About Us

peopleHum is a AI agentic Human Capital Platform, built for the next decade, on a mission to transform The Future of Work. Winner of the 2019 Global Codie Award and used by organizations around the world.

peopleHum is an intuitive, agile and integrated platform built for a complete multi-generational employee experience from hiring to performance, engagement and HRMS powered by machine learning and automation. AI is now central to how our product works, and the infrastructure behind it is one of the most interesting problems on our engineering roadmap.

Find out more: https://www.peoplehum.com

The Role

As a DevOps Intern, you'll work alongside our platform team on the systems that keep peopleHum and internal agentic meshes running and on the newer infrastructure that powers our AI features. You will also work on upgrading, managing and supporting an AI lab with blade servers and GPU machines like NVDIA Blackwell, AMD Ryzen, Apple silicon to run LLM models. This is a chance to learn classical DevOps and the emerging MLOps stack at the same time, which is a rare combination this early in a career.

What You'll Do

Core DevOps

• Assist in building and maintaining CI/CD pipelines using GitHub Actions, Jenkins, GitLab CI, or similar.
• Support infrastructure provisioning using Infrastructure as Code tools like Terraform or AWS CloudFormation.
• Work with containerization and orchestration technologies including Docker and Kubernetes.
• Help monitor and improve system performance, reliability, and scalability using Prometheus, Grafana, or the ELK stack.
• Write scripts (Bash, Python, or similar) to automate operations tasks and improve workflows.
• Collaborate with developers to understand deployment needs and automate routine work.
  
  

AI & ML Infrastructure

• Help deploy and operate machine learning and LLM-powered services, model serving endpoints, inference APIs, and the pipelines that ship them.
• Support the data plumbing behind our AI features: embedding pipelines, vector database operations, and data versioning.
• Build observability for AI workloads tracking inference latency, token consumption, cost per request, and failure modes that don't show up on a standard infra dashboard.
• Assist with GPU and accelerated compute provisioning, autoscaling, and cost optimization across cloud environments.
• Help automate model deployment workflows: versioning, staged rollouts, rollback, and reproducible environments for ML services.
• Explore AI-assisted tooling in our own engineering workflow using coding agents and automation to reduce toil in the deployment and incident response cycle.
  
  

Team

• Participate in code reviews, stand-ups, and knowledge-sharing sessions with senior DevOps and platform engineers.
  
  

What You'll Learn

• Real-world DevOps practices in a production-grade software environment.
• Hands-on experience with AWS, Azure, or GCP cloud platforms.
• The fundamentals of CI/CD, containerization, and microservice deployment.
• How AI systems actually run in production model serving, inference cost management, and the ways ML deployment differs from ordinary application deployment.
• How large scale systems are monitored, secured, and maintained.
• Agile development and collaboration in a cross functional team.
  
  

Eligibility

Final-year student or recent graduate in B.E./B.Tech (Computer Science, IT, or a related field). We care more about demonstrated curiosity and a willingness to learn than about a checklist of tools if you've been teaching yourself this stack, tell us what you've built.

Must have skills

• Good understanding of Linux/Unix systems and comfort with basic shell scripting.
• Familiarity with Git and version control workflows.
• Working knowledge of at least one cloud platform (AWS, Azure, or GCP).
• Exposure to Docker, or a solid grasp of what containers are and why they matter.
• Scripting ability in Python or Bash.
  
  

Good to have skills

• Exposure to CI/CD tools or pipeline concepts.
• Any hands-on experience with Kubernetes.
• Familiarity with Infrastructure as Code (Terraform, CloudFormation, Pulumi).
• Awareness of MLOps tooling such as MLflow, Weights & Biases, or model registries.
• Interest in observability and monitoring tooling.