Role Overview

Own and drive end-to-end enterprise data architecture on Google Cloud, covering lakehouse design, data engineering, graph-based data modeling, enterprise knowledge governance and AI-ready data platforms. Lead architecture decisions across data ingestion, processing, storage, analytics, knowledge management and agentic AI capabilities.

Required Technical Skills

- GCP Data Platform: BigQuery, BigLake, Cloud Storage, Cloud Spanner, BigQuery Graph.
- Data Engineering: Dataflow, Apache Beam, Pub/Sub, Dataproc, Cloud Composer, Dataform, STS.
- Lakehouse: Apache Iceberg, Parquet, open table formats and cross-cloud data integration.
- Governance & Knowledge: Dataplex, Knowledge Catalog, KAG, metadata management, ontology, lineage and data quality.
- Graph & Semantic Modeling: Cloud Spanner Graph, GQL, BigQuery Graph, graph modeling, semantic layers and reusable Measures.
- AI & GenAI: Vertex AI, Gemini, Model Garden, Vertex AI Search, Agent Runtime, ADK, A2A, MCP, and RAG / GraphRAG architectures.
- AI Retrieval & Embeddings: Vertex AI Embeddings, BigQuery Vector Search, Vertex AI Vector Search, hybrid search, re-ranking, chunking strategies and Document AI. • AI Governance & Evaluation: Responsible AI, Model Armor, Sensitive Data Protection (DLP), Gen AI Evaluation, prompt and agent testing, observability and LLMOps / AgentOps on Vertex AI Pipelines.
- Networking & Security: VPC, Shared VPC, VPC Service Controls, Private Service Connect, Cloud NAT, IAM, Cloud KMS and Secret Manager.
- DevOps & Operations: Terraform, Cloud Build, Artifact Registry, Cloud Logging and Cloud Monitoring.
- Architecture: Enterprise data modeling, dimensional modeling, distributed data architectures and cloud modernization.

Qualifications

- 10+ years of experience in enterprise data architecture and solution design. • Minimum 3+ years of hands-on GCP architecture experience.
- Proven experience designing enterprise-scale data platforms and cloud modernization solutions.
- Strong experience in data governance, graph/semantic modeling and enterprise knowledge architecture.
- Demonstrated ability to lead architecture decisions and cross-functional engineering teams.
- Google Professional Data Engineer or Professional Cloud Architect certification preferred.

Positive to Have

- Experience with enterprise knowledge graphs, ontology and semantic modeling, Neo4j. • Exposure to MCP, agentic frameworks and graph-augmented retrieval. • Experience with enterprise governance, Legal, Risk and Compliance data platforms. • Experience designing secure, scalable and AI-ready enterprise data ecosystems. • Experience taking GenAI and agentic solutions from PoC to production, including evaluation frameworks and cost / latency tuning.
- Familiarity with multi-agent frameworks (ADK, LangGraph, LangChain) and agent interoperability standards such as A2A and MCP.
- Hands-on exposure to model fine-tuning, context caching and Responsible AI / AI risk-management frameworks.

Key Responsibilities

- Enterprise Data Architecture
- Define end-to-end target architecture across GCP data, analytics and AI services. • Architect lakehouse solutions using Cloud Storage, BigLake, BigQuery and Apache Iceberg.
- Define integration patterns across AWS and GCP, including cross-cloud data movement and hybrid connectivity.




- Lead architecture discovery, source-system assessment, data readiness and technology selection.
- Data Engineering & Modeling
- Design Bronze/Silver/Gold medallion architectures and enterprise dimensional data models.
- Define fact, dimension and reference table structures, relationships, primary keys and cardinality.
- Establish ingestion and processing patterns using Storage Transfer Service (STS), Pub/Sub, Dataflow, Dataproc and Cloud Run.
- Define transformation and orchestration strategies using Dataform, BigQuery SQL and Cloud Composer (Airflow).
- Establish BigQuery partitioning, clustering, performance optimization and cost management strategies.
- Graph & Semantic Architecture
- Design graph-based data models using BigQuery Graph and Cloud Spanner Graph (GQL).
- Define graph nodes, edges, typed relationships, cardinality and graph update strategies.
- Architect reusable Measures and semantic models for consistent KPI calculations and business definitions.
- Define graph integration patterns to support knowledge discovery, contextual retrieval and AI-driven querying.
- Establish ontology and semantic modeling standards for enterprise knowledge representation.
- Enterprise Data Governance, Knowledge Catalog & KAG
- Design enterprise metadata and knowledge architecture using Dataplex, Knowledge Catalog and Google Cloud Knowledge Graph (KAG) capabilities.
- Define business glossary, technical metadata, asset relationships, lineage, classification and data quality frameworks.
- Design knowledge representation models, ontology structures, entities, relationships and semantic mappings across enterprise data sources.
- Establish integration patterns between structured data, unstructured documents, enterprise Systems of Record and knowledge graphs.
- Define metadata-driven knowledge discovery, graph construction, lifecycle management and governance workflows.
- Establish stewardship, metadata validation, approval and version management processes.
- Enable governed knowledge retrieval and contextual grounding for enterprise AI agents.
- AI & Knowledge Architecture
- Design AI-ready data foundations for Conversational Analytics, RAG and agentic AI. • Define integration patterns across Vertex AI, Gemini, Vertex AI Search and Agent Runtime/ADK.
- Architect knowledge graph and retrieval patterns integrating structured data, unstructured documents and disconnected enterprise Systems of Record. • Establish metadata-driven grounding, traceability and governance principles for AI agents.
- Define architecture patterns for MCP-based integrations and graph-augmented retrieval. • Define the reference architecture for RAG pipelines: document ingestion and parsing, chunking strategies, embedding generation (Vertex AI Embeddings), vector indexing (BigQuery Vector Search, Vertex AI Vector Search, AlloyDB pgvector) and hybrid (semantic + keyword) retrieval with re-ranking.




- Architect GraphRAG / graph-augmented retrieval combining knowledge graphs (BigQuery Graph, Spanner Graph, Neo4j) with vector search to support multi-hop reasoning and entity-aware answers.
- Design Conversational Analytics and natural-language-to-SQL patterns on BigQuery, including semantic layer grounding, curated metrics and Measures, and guardrails for query accuracy and cost.
- Define agentic AI architectures using Agent Development Kit (ADK) and Agent Runtime: single- and multi-agent orchestration, tool and function calling, planning and memory, agent-to-agent (A2A) communication, and human-in-the-loop approval for sensitive actions.
- Architect MCP server and tool integrations that expose governed enterprise data, Systems of Record and APIs to agents securely, with scoped permissions and full auditability.
- Establish the model strategy and selection framework across Gemini and other foundation models on Vertex AI (Model Garden): model fit, latency and cost trade-offs, prompt management, context caching, and fine-tuning versus grounding decisions.
- Define AI-ready data practices for unstructured content using Document AI and Vertex AI Search: OCR and layout parsing, entity and relationship extraction, metadata enrichment, and automated knowledge graph population from documents.
- Design AI governance and Responsible AI controls: grounding and citation of sources, data lineage into AI outputs, access-aware retrieval (row, column and document-level security), PII protection with Sensitive Data Protection (DLP), prompt injection defences, Model Armor and content safety filters.
- Establish evaluation and observability for AI systems: golden datasets, Vertex AI Gen AI Evaluation, groundedness, relevance and hallucination metrics, regression testing of prompts and agents, tracing, token and cost monitoring, and feedback loops.
- Define MLOps / LLMOps and AgentOps practices: CI/CD for prompts, pipelines and agents, Vertex AI Pipelines, model and prompt versioning, staged rollouts, and production monitoring with drift and quality alerts.
- Architect production serving for AI workloads covering scalability, latency SLAs, caching, quota and capacity management (including provisioned throughput), resiliency and cost optimization.
- Partner with business, data science and security stakeholders to prioritise AI use cases, define success metrics and deliver architecture roadmaps from PoC to production.
- Security & Infrastructure Architecture
- Define enterprise security architecture covering IAM, service accounts, Cloud KMS, Secret Manager and VPC Service Controls.
- Architect VPC, Shared VPC, Cloud NAT, Private Google Access, Private Service Connect and hybrid connectivity.
- Define secure data access, encryption, audit logging and compliance requirements. • Collaborate with DevOps teams on Terraform-based infrastructure provisioning and CI/CD architecture.
- Architecture Governance & Leadership
- Lead architecture reviews, technical design discussions and engineering teams. • Define architectural standards for scalability, reliability, performance and cost optimization.
- Review source-to-target data parity and ensure alignment with enterprise architecture. • Own architecture documentation, technical decisions, risk assessment and final design sign-off.

📌 Senior Data Architect (Bengaluru)
🏢 Niveus
📍 Bengaluru