DevOps Engineer

The core responsibilities for the job include the following:

Cloud Infrastructure Ownership:

• Own the design, evolution and day-to-day operation of cloud infrastructure on GCP.
• Manage multi-environment infrastructure (dev/staging/prod/preview/playground) with clear isolation and promotion paths.
• Build and maintain Infrastructure-as-Code, with reusable modules and best practices.
• Ensure high availability, scalability, backups, disaster recovery and secure access controls.
• Act as a point of escalation for infrastructure and platform-related issues.

Backend and Data Ops:

• Operate and support production microservices running on Kubernetes (GKE).
• Support the reliability and scalability of data systems and pipelines from an infra/ops standpoint.
• Own operational aspects of databases and background jobs, queues and workflows.
• Partner with backend and data engineers to improve performance, failure isolation, and observability with ease of debugging in production.

Release Engineering:

• Design, maintain and evolve CI/CD pipelines.
• Enable one-click deployments, safe release strategies (blue-green, canary, rollback) and environment promotion workflows.
• Build and maintain ephemeral preview environments per PR.
• Improve developer experience through standardised service templates, opinionated base images and self-service infrastructure tooling.
• Continuously reduce deployment friction and operational toil.

Observability:

• Own end-to-end observability across services using OpenTelemetry.
• Design and maintain dashboards, alerts and on-call hygiene.
• Define and track SLIs, SLOs, SLAs and error budgets.
• Lead incident response, postmortems and reliability improvements.
• Ensure all systems follow 12-factor app principles and are debuggable by default.

Security and Compliance:

• Apply security-by-default and least-privilege principles across cloud, Kubernetes and CI/CD.
• Own and manage IAM, secrets management and access controls for production systems.
• Ensure encryption in transit and at rest for services, databases and data pipelines.
• Secure Kubernetes workloads using RBAC, namespace isolation, network policies, and container best practices.
• Integrate vulnerability scanning and security checks into CI/CD pipelines.
• Ensure PHI/PII is handled safely across applications, data platforms and AI systems.
• Maintain audit trails and traceability for production changes and access.
• Support infrastructure practices aligned with HIPAA and SOC2 security requirements.

MLOps and LLMOps:

• Enable production deployment of fine-tuned LLM models.
• Support GPU-backed inference workloads via GPU node pools, autoscaling and cost-aware scheduling.
• Help standardise patterns for model serving, model versioning, rollout strategies and canary or shadow deployments for models.
• Ensure AI services have proper observability for latency, throughput, error rates and cost per request.
• Drive FinOps practices, including compute and GPU cost optimisation, right-sizing, storage and network efficiency.

Requirements:

• 3+ years of hands-on experience in DevOps/Platform/SRE roles.
• Strong, production-level experience with GCP (Compute, IAM, VPC, Cloud Run, etc. ).
• Deep experience with Docker and Kubernetes (GKE).
• Proven ownership of CI/CD pipelines (GitLab CI + ArgoCD) and release workflows.
• Experience with Infrastructure-as-Code (Terraform) and service mesh (Istio).
• Experience with OpenTelemetry and tools like Signoz, Prometheus, Grafana, ELK, Loki, etc.
• Strong Linux fundamentals and scripting skills (Bash, Python).
• Experience operating production systems with uptime and latency requirements.
• Strong debugging, incident management and root cause analysis skills.
• Ability to work independently and make sound technical decisions.

Bonus:

• Exposure to ML infrastructure, MLOps or LLMOps.
• Experience running GPU workloads or inference services.
• Familiarity with data platforms or workflow engines.
• Experience in regulated domains (healthcare, fintech, etc. ).

Skills

GCP, GKE, Google Cloud, Kubernetes, Terraform