Country/Region:
IN
Requisition ID:
34644
Work Model:
Position Type:
Salary Range:
Location:
INDIA - PUNE - BIRLASOFT OFFICE - HINJAWADI
Title: Architect
Description:
Area(s) of responsibility
Roles & Responsibilities
- Architect and design Azure-based data lake/lakehouse platforms, domain data models, and ingestion-to-consumption pipelines.
- Develop conceptual, logical, and physical cloud data models aligned with enterprise standards.
- Architect RAG pipelines including embeddings, chunking, vector stores, hybrid retrieval, reranking, and evaluation.
- Build Agentic AI workflows using LangChain and LangGraph; design tool orchestration, memory, and safety layers.
- Implement governance with Microsoft Purview for cataloging, lineage, PII tagging, and policy enforcement.
- Ensure platform security using Entra ID, private endpoints, VNETs, Key Vault, and encryption controls.
- Lead solution architecture reviews, performance tuning, cost optimization, and NFR engineering.
- Oversee CI/CD (Azure DevOps), IaC (Terraform/Bicep), and observability (Azure Monitor, App Insights).
- Mentor engineering teams and standardize best practices, patterns, and reusable components.
Technical Skills
Mandatory
- Azure Data Platform: ADLS Gen2, Synapse/Serverless SQL, Databricks/Spark, ADF/Synapse Pipelines
- Programming: Python, PySpark, SQL
- GenAI & Agentic AI: RAG architecture, vector stores (Azure Cognitive Search, Pinecone, Weaviate, Qdrant), embeddings, reranking
- Frameworks: LangChain, LangGraph
- Data Modeling: Conceptual/logical/physical models, Delta/Parquet patterns, lakehouse modeling
- Data Governance: Microsoft Purview (catalog, lineage, classification, glossary, PII governance)
- Security: Entra ID, RBAC/ABAC, Key Vault, VNET integration, encryption
- SDLC & DevOps: Azure DevOps (CI/CD), Terraform/Bicep, ADRs, HLD/LLD documentation
- Performance & Cost Optimization across compute, storage, vector workloads, and pipelines