F2F -Hiring Alert ! EPAM ! GCP Data Engineer - Hyderabad!

Epam Systems; Hyderabad, Chennai, Bengaluru; On-site; INR 1,500,000 - 2,200,000; Full time

Job summary

Epam Systems is seeking a senior data engineering professional to design and implement a robust data lake architecture across GCP BigQuery, Iceberg, and Delta Lake ecosystems. The role focuses on building scalable lakehouse pipelines with CDC via Kafka for real-time data in a multi-product environment.

You will validate Snowflake Horizon catalog definitions and Delta Sharing/Unity Catalog integrations, ensuring secure, governed access and high data freshness across Databricks consumers.

Qualifications

• Experience with cloud-based data lakes and lakehouse architectures.
• Hands-on experience with streaming and CDC patterns (Kafka).
• Proficiency in Python and Spark for data processing.
• Familiarity with Snowflake Horizon catalog and Databricks Unity Catalog.
• Knowledge of Iceberg and Delta Lake formats for data storage.

Responsibilities

• Design and implement lakehouse with Delta and Iceberg metadata.
• Build robust GCS-to-BigQuery ingestion patterns for operational data.
• Implement CDC and real-time data movement via Kafka.
• Develop data lineage, bookkeeping, and reusable adapters across pipelines.
• Ensure modular, version-controlled data integration code.

Skills

GCP BigQuery

Python

Spark

Kafka/CDC

data lake architecture

Tools

Apache Iceberg

Delta Lake

DeltaSharing

Iceberg UniForm

Snowflake Horizon catalog

Databricks Unity Catalog

Job description

Role Need

Location : All India EPAM

Interview Mode -Hyderabad Face to Face -17oct-26 - Saturday .

Years of Experience -5+years.

Required Tech Skills

• GCP BigQuery (advanced)
• Apache Iceberg
• Delta Lake
• DeltaSharing
• Kafka/CDC
• Python
• Spark
• data lake architecture
• Iceberg UniForm
• Snowflake Horizon catalog
• DatabricksUnity Catalog

AI Tooling Proficiency

Operative actively uses AI-assisted development tools across its engineering workflow. All resources are expected to have active, working experience with AI coding and productivity tools. Claude Code (Anthropic) is the preferred tool; demonstrated proficiency with GitHub Copilot, Cursor, or comparable tools is acceptable. Candidates should be prepared to demonstrate AI tooling use in a technical screening.

Responsibilities

Data Lake Architecture

• Design and implement lakehouse UniForm write layer one physical dataset, dual-format metadata (Delta +Iceberg), readable by all target consumers without conversion
• Establish and validate GCS-to-BigQuery ingestion pipeline patterns for structured operational data types (sales,delivery, schedule, performance)
• Implement change-data-capture (CDC) patterns using Kafka for real-time and near-real-time data movement into thelakehouse
• Develop dependency-aware bookkeeping and data lineage tracking patterns for use across all data pipelines
• Ensure all adapter code is modular, version-controlled, and designed for reuse across new data source integrations

Snowflake Integration

• Configure Iceberg external table definitions for Operative data products within Snowflake's Horizon catalog
• Validate zero-copy read access from Snowflake to Operative-managed Iceberg / Delta tables without data movement
• Implement and test tenant-scoped access controls compatible with Snowflake Horizon catalog metadata governance
• Document integration pattern for re-use across additional data products

Databricks Integration

• Implement and certify Delta Sharing adapter for live, zero-copy data sharing from Operative's Delta Lake tables to Databricks consumers
• Configure Delta Sharing endpoint registration and sharing agreement management
• Validate read access from Databricks via Delta Sharing protocol — Pandas, Spark, and other Delta-compatible consumers
• Test end-to-end data freshness and sharing latency within agreed SLA targets

Governed Access Layer (Integration Exchange)

Operative operates a governed data access layer that controls all external data-out connectivity. This workstream requiresthe SI to implement connector registry entries, RBAC configurations, and tenant-scoped authorization for both the

• Snowflake and Databricks as named connector types in the connector registry
• Implement RBAC and tenant-scoped authorization on all data-out paths — auditable and controlled per tenant
• Implement metering hooks compatible with the billing framework for governed data-out flows

Validate authori

Warm Regards

Kalaichandran

[e-mail]