GM - IT Analytics

Vodafone Idea is looking for a seasoned AI Architect to design, build, and scale enterprise-grade AI and Generative AI (GenAI) solutions. In this role, you will be responsible for defining the technical architecture and roadmap for AI applications deployed on top of our enterprise Cloudera Data Platform (CDP 7.1.9 / 7.3.1) .

You will bridge the gap between complex big data ecosystems and cutting-edge machine learning/LLM workflows. The ideal candidate brings hands-on expertise in model selection, lifecycle governance, prompt engineering, and production-grade MLOps/AIOps integrated into modern DevOps pipelines.

Lead the end-to-end architecture and integration of Predictive AI and Generative AI applications built directly on CDP 7.1.9 / 7.3.1 infrastructure.

Design high-throughput, low-latency data pipelines and feature stores leveraging CDP components (Spark, Hive, Impala, Kafka, Apache Iceberg).

Establish retrieval-augmented generation (RAG) frameworks, fine-tuning setups, and vector search systems connected to enterprise big data repositories.

Evaluate, benchmark, and select optimal foundational models (LLMs, SLMs, open-source vs. proprietary) and classic ML algorithms tailored for telecom use cases (e.g., customer churn, network optimization, hyper-personalization, fraud detection).

Enforce robust AI Governance frameworks covering model drift detection, explainability (XAI), fairness/bias mitigation, data privacy, and compliance.

Manage model registry, lifecycle versioning, and policy enforcement across dynamic runtime environments.

Establish standardized prompt engineering methodologies, context-window management, and agentic workflows (e.g., LangChain, LlamaIndex, AutoGen).

Optimize token utilization, latency, throughput, and inference costs for LLM deployments.

Implement guardrails, output validation, and safety filters to prevent hallucination, data leakage, and security risks in public/private LLM interactions.

Design and implement robust MLOps/AIOps pipelines for automated model training, continuous integration, testing, deployment, and real-time monitoring (CI/CD/CT).

Partner with DevOps teams to deploy containerized AI microservices (Kubernetes, Docker) within hybrid-cloud and on-premises infrastructure.

Configure comprehensive telemetry, logging, and observability tools (Prometheus, Grafana, MLflow, Weights & Biases) to maintain continuous operational health.