Data Scientist

CloudXtreme Technology Private Ltd; Bengaluru, Mumbai; On-site; INR 1,800,000 - 3,000,000; Full time

Job summary

CloudXtreme Technology Private Ltd. is seeking an ML Infra Lead to design and manage scalable deployment pipelines for ML models. You will build and maintain infrastructure with Kubernetes, Docker, and cloud services, ensuring reproducibility and robust observability.

You will collaborate with ML Engineers and Data Scientists, optimize costs, and implement secure, compliant deployment practices, including GPU/accelerator management for LLM workloads.

Qualifications

• B.E./B.Tech degree is required.
• Mandatory certification in Machine Learning or Data Science.

Responsibilities

• Design and manage CI/CD pipelines for ML model deployment.
• Provision and maintain scalable infra using Kubernetes, Docker, AWS Lambda, or cloud services (AWS/GCP/Azure).
• Automate model deployment workflows with reproducibility and version control.
• Implement monitoring, alerting, and observability for ML systems (latency, throughput, data drift, failures).
• Integrate with model orchestration pipelines (Airflow, Prefect) to support retraining and inference workflows.
• Manage model registry, versioning, and rollout strategies (blue/green, canary deployments).
• Ensure security, compliance, and cost optimization for ML infra.
• Manage vector databases (OpenSearch, Pinecone, FAISS, Milvus) for GenAI retrieval pipelines.
• Work closely with ML Engineers and Data Scientists to understand model requirements.
• Support real-time and batch inference pipelines by ensuring infra scalability and resilience.
• Troubleshoot deployment and runtime issues across containers, APIs, and cloud services.
• Document and standardize infrastructure-as-code practices (Terraform, Helm, etc.).
• Manage GPU/accelerator infra for LLM fine-tuning or inference optimization.

Skills

Leadership

Team management

CI/CD

Monitoring

Observability

Security

Cost optimization

Education

B.E./B.Tech

Tools

Kubernetes

Docker

Terraform

Airflow

Prefect

Helm

OpenSearch

Pinecone

FAISS

Milvus

AWS

GCP

Azure

Job description

Key / Primary Responsibilities: '

• Design and manage CI/CD pipelines for ML model deployment.


• Provision and maintain scalable infrastructure using Kubernetes, Docker, AWS Lambda, or cloud services (AWS/GCP/Azure).


• Automate model deployment workflows, ensuring reproducibility and version control.


• Implement monitoring, alerting, and observability for ML systems (latency, throughput, data drift, failures).


• Integrate with model orchestration pipelines (Airflow, Prefect) to support retraining and inference workflows.


• Manage model registry, versioning, and rollout strategies (blue/green, canary deployments).


• Ensure security, compliance, and cost optimization for ML infra.


• Manage vector databases (Opensearch, Pinecone, FAISS, Milvus) for GenAI retrieval pipelines.

Secondary Responsibilities:

• Work closely with ML Engineers and Data Scientists to understand model requirements.


• Support real-time and batch inference pipelines by ensuring infra scalability and resilience.


• Troubleshoot deployment and runtime issues across containers, APIs, and cloud services.


• Document and standardize infrastructure-as-code practices (Terraform, Helm, etc.).


• Manage GPU/accelerator infra for LLM fine-tuning or inference optimization.

Key Success Metrics: Model Uptime and Reliability, Model Performance in production, Efficiency of Spark Jobs, Workflow Automation Efficiency, User Satisfaction and Internal Feedback.

Professional Degree: Any B.E/B.Tech

Mandatory Certification: Machine Learning or Data Science.

Managerial & Leadership Responsibilities: 'Leading a team of 4 Data scientist and ML engineers to deliver on primary and secondary responsiblities..