Role Overview
DevOps/SRE Engineer responsible for building, automating, securing, and operating the cloud platform that supports AI/ML services. Owns infrastructure-as-code, CI/CD, indexing-job orchestration, observability, access controls, incident response, cost optimization, and reliability across development, UAT, and production environments.
Responsibilities:
• Build and operate the AWS platform that hosts real-time/batch prediction and ranking services.
• Provision and maintain private networking, compute, API endpoints, IAM roles, DNS, encryption, secrets, logging, and monitoring through Terraform.
• Maintain separate development, test, UAT, staging, and production environments with consistent configuration, controlled access, and repeatable deployments.
• Support managed AWS services used by ML and retrieval workloads, such as Lambda, Step Functions, ECS/Fargate, EKS, API Gateway, S3, DynamoDB, SageMaker, OpenSearch, CloudWatch, and Secrets Manager.
• Would be responsible for writing the data indexing pipelines and design and develop API's
• Partner with data scientists and application teams to make the ML services/data pipelines deployable, observable, scalable, and supportable in production.
Core Responsibilities:
Own the full infrastructure footprint in Terraform (or equivalent): private networking, compute, managed ML/search services, API gateway, IAM, encryption, DNS, and logging.
Manage multi-account, multi-environment deployments through CI/CD with automated, semantic versioning and disciplined commit conventions.
Keep infrastructure reviewable, reproducible, and debuggable; resolve networking, access, and build-pipeline issues.
Security & authentication
Implement and operate service-to-service authentication (OAuth2/JWT, API authorizers with role-based access), manage secret rotation, and own auth-related incident response.
Apply least-privilege access by default and flag any unauthenticated or over-permissioned surface before it ships.
Operations, observability & incident response
Investigate production incidents from logs and metrics across the stack, and distinguish client-side from backend root causes with evidence.
Run load and capacity tests to establish real limits and maintain a mitigation playbook for traffic spikes.
Design cost-aware performance strategies (scale-to-zero vs. always-warm vs. warmers) and tune the latency/cost tradeoff.
Cross-team collaboration & communication
Required Qualifications
Deep hands-on AWS (or comparable cloud), especially serverless compute.
Production experience with infrastructure-as-code in a multi-account/multi-environment setup.
Solid CI/CD, Git hygiene, and automated testing discipline.
Strong Python and comfort owning production services end to end.
Strong operational instincts: reading logs and metrics, incident triage, and reasoning about latency, cost, and blast radius.
Clear written communication across technical and non-technical audiences
