DevOps Manager

Position: DevOps ManagerLocation: India (Remote)Employment Type: Full-TimeSchedule: Monday to Friday, Day ShiftCompany DescriptionScry AI is a research-led enterprise AI company that builds intelligent platforms to drive efficiency, insight, and compliance. Our platforms Collatio®, Auriga®, and Concentio® streamline complex workflows by automating data extraction, validation, reconciliation and delivering real-time intelligence.We are seeking a DevOps Manager to lead our infrastructure, CI/CD, and reliability practices across cloud and on-prem deployments. You will own uptime, performance, security, and cost efficiency for AI/ML workloads powering Collatio®, Auriga®, and Concentio®.Role OverviewAs DevOps Manager, you will lead a small team of DevOps/SRE engineers to design, automate, and operate secure, compliant, and highly available platforms across AWS/Azure/GCP and customer on-prem environments. You will standardize IaC, improve CI/CD velocity, build robust observability, and enable GPU-accelerated AI inference at scale for enterprise clients.Key ResponsibilitiesPlatform Reliability & Operations• Own SLOs/SLIs, availability, latency, and capacity planning across services.• Lead incident response, root-cause analysis, postmortems, and on-call processes.• Implement backup, disaster recovery, and business continuity for multi-region and on-prem.Cloud, On-Prem & Edge Deployments• Architect Kubernetes platforms (managed and self-hosted), including RBAC, network policies, and secrets management.• Standardize infrastructure with Terraform, Helm, and GitOps (Argo CD) for repeatable customer deployments.• Support Concentio® edge/IoT rollouts with secure remote updates and telemetry pipelines.AI/ML & Data Infrastructure• Enable GPU scheduling and drivers (CUDA, NVIDIA), inference runtimes (Triton), and model packaging.• Build MLOps foundations (MLflow, feature stores) and artifact/version governance.• Operate data services (Kafka, PostgreSQL, Redis, MinIO/S3, Elasticsearch/Opensearch) for high-throughput pipelines.CI/CD & Developer Experience• Own CI/CD with GitHub Actions/GitLab CI/Jenkins; establish trunk-based development, automated testing, and canary/blue-green releases.• Maintain internal developer platforms, templates, and golden paths to improve delivery speed and quality.Security, Compliance & Observability• Implement least-privilege access, SSO (Okta/AAD), Vault-based secrets, image scanning (Trivy), and policy as code.• Ensure SOC 2, ISO 27001, HIPAA/GDPR alignment with audit trails and immutable logs.• Build end-to-end observability using Prometheus, Grafana, Loki/EFK, and OpenTelemetry.FinOps & Stakeholder Management• Track cloud spend, rightsize resources, and negotiate quotas for GPU/compute.• Partner with Product, Data Science, and Customer Success to plan capacity for new features and enterprise go-lives.Required Qualifications & Skills• Strong Kubernetes expertise (production operations, networking, security, Helm, GitOps).• Proven IaC experience with Terraform and configuration management (Ansible).• CI/CD at scale with GitHub Actions/GitLab CI/Jenkins; artifact registries and SBOMs.• Observability: Prometheus, Grafana, ELK/EFK or Loki, alerting and runbooks.• Cloud proficiency in at least one major provider (AWS/Azure/GCP) and Linux fundamentals.• Security fundamentals: network segmentation, TLS, secrets management, container hardening.• Experience running data/streaming systems (Kafka, Redis, PostgreSQL) in production.• Excellent communication, incident leadership, and stakeholder management.Nice-to-Have• GPU orchestration, Triton Inference Server, Hugging Face model serving.• Service mesh (Istio/Linkerd), API gateways, and zero-trust patterns.• MLOps tooling (MLflow, Feast), Airflow, dbt.• Compliance implementations for regulated industries (BFSI, healthcare).• Certifications: CKA/CKAD, AWS/Azure/GCP Architect, Security+.Our Ideal Candidate• Drives reliability with automation, not toil.• Balances speed and safety with measurable delivery improvements.• Thrives in customer-facing, hybrid cloud, and on-prem environments.• Coaches teams with clear standards, runbooks, and continuous improvement.Tip for CandidatesIf you want to build secure, high-performance platforms for real-world AI at enterprise scale, follow our page for more such relevant job openings.