Senior Data Engineer

Peak Finder Consulting Chennai; Chennai District, Bengaluru; Hybrid; INR 700,000 - 1,100,000; Full time

Job summary

Peak Finder Consulting Chennai is seeking an experienced engineer to architect and scale a low-latency analytics stack for thousands of SaaS users. You will build real-time ingestion from PostgreSQL and streaming sources into Apache Pinot and model metrics with Cube.js in a semantic layer.

You will optimize Pinot performance, expose data via REST/GraphQL APIs, and implement multi-tenant data governance to ensure strict data isolation across customers.

Qualifications

• Architect and scale a low-latency analytics stack for thousands of SaaS users.
• Ingest data in real-time from PostgreSQL and event streams (Kafka/Kinesis) into Pinot.
• Model metrics using Cube.js in a semantic layer for consistent reporting.
• Optimize Pinot tables, indexes and pre-aggregations for fast dashboards (<300ms).
• Expose data models via REST/GraphQL APIs for frontend visualizations.
• Implement multi-tenant data governance to isolate customer data.

Responsibilities

• Design and maintain a scalable analytics stack for high-concurrency SaaS queries.
• Build and optimize real-time data ingestion pipelines from transactional DBs and event streams.
• Develop a semantic layer with Cube.js to model sales metrics.
• Tune Pinot for performance and implement effective pre-aggregations.
• Provide REST/GraphQL APIs for frontend consumption and visualizations.
• Enforce data isolation and governance across tenants.

Skills

Analytics architecture

Real-time ingestion

Semantic layer design

Performance optimization

API integration

Data governance

Tools

Apache Pinot

Cube.js

PostgreSQL

Kafka

Kinesis

REST/GraphQL

Job description

Key Responsibilities:

• Architecture & Scaling: Design and maintain a low-latency analytics stack capable of handling high-concurrency queries from thousands of concurrent SaaS users.
• Real-time Ingestion:Build and optimize ingestion pipelines that move data from transactional databases (PostgreSQL) and event streams (Kafka/Kinesis) into Apache Pinot.
• Semantic Layer Development:UseCube (Cube.js)to model complex sales metrics (e.g., "Sequence Conversion Rate," "Attributed Revenue," "Meeting Booked Rate") ensuring consistency across the application.
• Performance Engineering:OptimizeApache Pinottables, indexing strategies, andCubepre-aggregations to ensure dashboard widgets load in under 300ms.
• API Strategy:Expose data models via REST/GraphQL APIs, collaborating closely with Frontend Engineers to build world‑class data visualizations.
• Data Governance:Implement multi‑tenant security logic within the semantic layer to ensure strict data isolation between different customer accounts.