Platform Engineer

Overview

Platform Engineer (MLOps & DataOps) on Priceline's MLOps & Data DevOps team, which builds and operates the infrastructure, automation, deployment pipelines, observability and self-service capabilities that let data engineering, machine learning and GenAI teams deliver workloads safely to production.

What you'll do

• Build, maintain and improve Terraform modules and configurations for GCP data and ML platform services.
• Provision, operate and automate Cloud Composer/Airflow, Dataproc, Dataflow, BigQuery, Cloud Storage and Vertex AI across development, QA and production environments.
• Support platform upgrades, configuration changes, capacity planning and operational improvements.
• Develop reusable automation and internal tools using Python and Shell scripting.
• Apply infrastructure-as-code, change-management, security and auditability standards to platform changes.
• Build and operate governed CI/CD pipelines for Airflow DAGs, Dataproc jobs, BigQuery changes, ML pipelines and model-serving infrastructure.
• Implement reliable deployment, promotion, validation, rollback and release processes using Git-based workflows.
• Work with GitHub Actions, Argo CD, Codefresh or equivalent CI/CD and GitOps tooling.
• Create self-service patterns that let data and ML teams provision and deploy workloads safely without bypassing governance controls.
• Improve developer experience, deployment reliability and time to production through standardization and automation.
• Support production ML workflows covering model training, pipeline execution, model deployment, endpoint operations, batch inference and retraining.
• Operate and improve ML platforms based on Vertex AI, Kubernetes/GKE, Docker, Kubeflow or equivalent technologies.
• Support model gateway and LLM platform capabilities such as LiteLLM, including model onboarding, access management, rate limiting, usage visibility, cost controls and reliability improvements.
• Help establish repeatable practices for model versioning, artifact management, model registries, feature stores and data/model lineage.
• Contribute to GenAI, RAG, model evaluation, guardrails and AI observability capabilities.
• Define and operate monitoring, logging, alerting and reliability practices for data pipelines, ML pipelines, model endpoints and platform services.
• Establish meaningful SLIs/SLOs, escalation paths, runbooks and error-management practices.
• Use tools such as PagerDuty, Splunk, New Relic, Arize or equivalent observability platforms.
• Troubleshoot pipeline failures, deployment issues, infrastructure problems, endpoint failures and performance degradation.
• Participate in on-call support, contribute to root-cause analysis and drive corrective and preventative actions.
• Integrate IAM, RBAC, service-account controls, vulnerability scanning, secrets management and compliance guardrails into platform workflows.
• Improve cloud cost visibility, resource utilization, capacity management and FinOps practices for data and ML workloads.
• Evaluate, adopt and operationalize platform capabilities that improve reliability, security, developer productivity and operational efficiency.
• Partner with Cloud Infrastructure, SRE, InfoSec, Data Engineering, Data Science and ML teams to resolve platform and operational issues.
• Document platform standards, operational procedures, troubleshooting guidance and reusable deployment patterns.

What you'll need

• B.Tech or B.E in Computer Science, Information Technology, or equivalent practical experience.
• 4–6 years of experience in DevOps, SRE, cloud platform engineering, MLOps, DataOps or a closely related discipline.
• Strong hands-on experience with Google Cloud Platform or equivalent public cloud technologies.
• Strong practical experience with Terraform for infrastructure provisioning, configuration management, reusable modules and controlled production changes.
• Hands-on experience with Docker and Kubernetes.
• Experience building or operating CI/CD pipelines and Git-based deployment workflows using GitHub Actions, Argo CD, Codefresh, Jenkins or equivalent tools.
• Proficiency in Python and Shell scripting for platform automation, operational tooling, pipeline hooks and internal utilities.
• Experience supporting Apache Airflow/Composer, Dataproc, Dataflow, Spark, BigQuery or comparable data-processing and workflow-orchestration technologies.
• Experience supporting ML pipelines, model deployments, inference endpoints, model-serving infrastructure or ML platform operations.
• Experience with monitoring, logging, alerting, incident response, on-call support, troubleshooting and root-cause analysis for production systems.
• Working knowledge of IAM, RBAC, service accounts, secrets, vulnerability management, networking fundamentals and secure cloud operations.
• Strong ownership, structured problem-solving, documentation, communication and collaboration skills.

Nice to have

• Experience operating LLM gateways or GenAI platforms such as LiteLLM, Tyk or equivalent model-access layers.
• Experience with ML/LLM observability and evaluation platforms such as Arize, LangSmith or equivalent tools.
• Experience with GKE platform operations, GPU workloads, vector databases, RAG infrastructure or model-serving systems.
• Experience implementing data, model, artifact or pipeline lineage and reproducibility practices.
• Experience with cloud cost optimization, capacity management and FinOps for data and ML workloads.
• Experience applying SRE, security, networking or compliance practices to cloud data and ML platforms.

Details

• Hybrid: two days onsite at the Mumbai office, as agreed with your manager (ideally Tuesday, Wednesday or Thursday); remaining days remote or in office.
• Location: Mumbai.