DevOps Engineer

Recrew AI; Bengaluru; On-site; INR 1,500,000 - 2,000,000; Full time

Benefits offered by this job

Competitive compensation

Access to cutting-edge GPU compute infrastructure

Opportunity to work alongside AI researchers

Job summary

Recrew AI is looking for a Cluster DevOps/SRE Engineer based in Bengaluru to design and manage GPU/CPU compute clusters on GCP. The role involves ensuring the reliability and performance of critical infrastructure for AI model training workloads.

Candidates should have 4–10 years of experience in DevOps or SRE roles and a strong background in managing large-scale GKE clusters. This position offers competitive compensation and an innovation-driven culture.

Qualifications

• 4–10 years of experience in DevOps, SRE, or infrastructure engineering roles.
• Strong experience managing GKE clusters at scale (100+ nodes).
• Proficiency in Python or Bash for automation.

Responsibilities

• Design and manage large-scale GKE clusters for AI/ML.
• Build CI/CD pipelines with Cloud Build and ArgoCD.
• Monitor cluster performance using Google Cloud Monitoring.

Skills

GKE (Google Kubernetes Engine)

Python

Bash scripting

CI/CD with ArgoCD

Terraform

Google Cloud Monitoring

Prometheus

Grafana

Tools

Google Cloud Platform

SLURM

Ray

Job description

Function: DevOps / Site Reliability Engineering / Infrastructure

Type: Full-time

Industry: Artificial Intelligence, Cloud Infrastructure, Telecommunications

About Company

The company is the dedicated AI research and innovation arm of a large-scale Indian telecom and technology conglomerate. It is building foundational AI technologies designed specifically for India's languages and digital economy.

Focus areas include multilingual speech recognition, voice synthesis, real-time conversational AI, and multimodal foundation models. The company collaborates with global AI leaders including OpenAI, Anthropic, Google, and Meta.

Its systems are built to serve hundreds of millions of users across India. It is investing across the full AI value chain — from next-generation data centers to consumer AI platforms and an agentic marketplace for SMEs.

Position Overview

The company is seeking a Cluster DevOps / SRE Engineer to design, manage, and optimize large-scale GPU/CPU compute clusters on GCP that power its AI research and model training workloads. The role owns the reliability, performance, and scalability of infrastructure critical to training and serving frontier AI models, sitting at the heart of the company's compute backbone.

Role & Responsibilities

• Design, deploy, and manage large-scale GKE (Google Kubernetes Engine) clusters for AI/ML training and inference workloads
• Build and maintain CI/CD pipelines using Cloud Build and ArgoCD to automate model training, evaluation, and deployment workflows
• Monitor cluster health, performance, and reliability using Google Cloud Monitoring and Grafana; drive incident response and root cause analysis (RCA)
• Optimize GPU utilization across multi-node training jobs on GCP, managing resource scheduling and job queuing systems
• Collaborate with AI researchers to provision GCP infrastructure for large-scale distributed training experiments
• Implement and enforce security, IAM access control, and compliance standards across GCP compute infrastructure
• Automate infrastructure provisioning and configuration management using Terraform on GCP

Must Have Criteria

• 4–10 years of experience in DevOps, SRE, or infrastructure engineering roles
• Strong hands‑on experience managing GKE (Google Kubernetes Engine) clusters at scale (100+ nodes)
• Proficiency in Python or Bash scripting for automation and infrastructure tooling
• Experience with CI/CD pipelines using ArgoCD or Cloud Build on GCP
• Hands‑on experience with Infrastructure-as-Code using Terraform on GCP
• Experience with observability and monitoring using Google Cloud Monitoring, Prometheus, and Grafana
• Familiarity with GPU cluster management and distributed computing frameworks such as SLURM or Ray on GCP

Nice to Have

• Experience managing AI/ML infrastructure for large-scale model training (LLMs, speech models) on GCP
• Prior work at an AI research lab, cloud provider, or high‑scale tech company
• Experience with GCP-native MLOps tooling such as Vertex AI Pipelines or Kubeflow on GKE
• Contributions to open‑source infrastructure or DevOps tooling projects

What We Offer

• Opportunity to build AI infrastructure that will serve hundreds of millions of users across India
• Work alongside world‑class AI researchers and engineers on frontier model development
• Access to cutting‑edge GPU compute infrastructure powered by GCP at scale
• Competitive compensation with the backing and scale of a major Indian conglomerate
• Fast‑paced, innovation‑first culture with direct impact on India's AI ecosystem