Bangalore, India Backend & platform engineering

Saravanan
Kumar.

Senior Backend Engineer · Distributed Systems & Data Platforms. I build reliable, high-volume data and search systems where cloud infrastructure and thoughtful engineering leadership meet.

View resume ↗ LinkedIn ↗ Get in touch ↘

Java Spring Boot Kafka OpenSearch Apache Spark OCI Kubernetes

Scroll to explore

01 / About

I build the systems that make complex data useful.

With 10+ years in production Java engineering across Oracle, Bosch, and WorksApplication, I take end-to-end ownership of backend platforms: from system design and data ingestion to production operations and continuous improvement.

My work centers on event-driven architectures, search and data platforms, Kafka pipelines, cloud-native services, and enterprise workflow automation. I pair hands-on technical design with incident ownership, code reviews, mentoring, and cross-functional delivery.

Distributed systems Data platforms Search at scale Technical leadership

02 / At a glance

10+

Years delivering production software

End-to-end

Ownership across event-driven platform delivery

High-volume

Data ingestion, indexing, and search workloads

Production

Reliability, recovery, performance, and safe deployment practices

Improved high-volume indexing throughput by addressing ingestion and search bottlenecks.

03 / Experience

Sep 2021 — Present Bangalore, India

Oracle India Private Ltd

Senior Software Engineer IC-III · Mar 2024 — Present

Click to collapse

Technical owner for a cloud-based, event-driven data and search platform supporting usage-consumption workflows, spanning ingestion, indexing, APIs, integrations, notifications, and applied-ML exploration.

Platform & architecture

Leads technical design across the platform and architecture reviews across teams.
Designed resilient monthly and rolling-three-month load orchestration with independent Spark extraction, sequential indexing, safe alias promotion, recovery controls, timestamp-based batch pairing, and protections against overlapping or out-of-order daily loads.
Built REST APIs for plan details, account-plan views, consumption trends, products, rate cards, saved searches, extension data, and chatbot-facing queries.

Scale & engineering leadership

Improved a Kafka, Spark, and OpenSearch pipeline through parallel processing and bulk-indexing practices for high-volume workloads.
Owns CI/CD, containerization, Kubernetes deployment strategy, incident response, and observability.
Drives cross-team delivery, mentors engineers, and conducts code reviews.

Java 17 Java 21 Spring Boot Kafka Spark OpenSearch OCI Kubernetes Oracle Database

Earlier role at Oracle Click to collapse

Senior Software Engineer IC-II · Sep 2021 — Feb 2024

Designed the initial materialized-view-based extraction and Elasticsearch indexing workflow, then contributed to its Spark-based evolution. Helped migrate Elasticsearch workloads to OpenSearch. Built asynchronous Kafka services and deployed containerized services to OCI-managed Kubernetes clusters.

Java 11 Java 17 Elasticsearch OpenSearch

Jan 2018 — Aug 2021 Coimbatore, India

BGSW formerly RBEI

Senior Software Engineer Product Development

Click to collapse

Built tooling for automotive firmware data preparation and integrity validation in embedded-system build workflows.

Selected contribution

Developed the HexmodX firmware console tool for HEX-file modification, CRC validation, and SHA integrity checks using Java security APIs.

Workflow improvement

Built an MDX-to-MSR conversion tool for embedded microcontroller build pipelines, streamlining source-data preparation for firmware builds.

Java Java Security APIs Embedded workflows

Jul 2016 — Dec 2017 Chennai, India

WAP formerly IVTL

Software Engineer Full Stack

Click to collapse

Delivered employee-management workflows for HUE, an enterprise ERP product used by Japanese companies.

Delivery

Delivered employee-management functionality from business requirements through implementation.

Search

Implemented Elasticsearch-backed employee search and maintained changes through Git-based development.

Java 8 Spring MVC JavaScript Elasticsearch Git

04 / Engineering work

From ingestion to action.

Selected initiatives drawn from enterprise platform, workflow, and applied-ML work.

Data & search platform

OCI Usage Consumption Platform

A cloud-based, event-driven platform for technology-consumption workflows. Work included data ingestion, indexing, search, integrations, notifications, shared data services, and applied-ML exploration.

Challenge Keep high-volume ingestion and search reliable across scheduled and incremental load scenarios.

Approach Parallel Kafka consumption and OpenSearch bulk indexing, Spark extraction, sequential indexing, safe alias promotion, and operational recovery controls.

Outcome Improved indexing throughput by resolving ingestion and search bottlenecks.

Kafka Apache Spark OpenSearch OCI Helm

Applied ML proof of concept

Extension Decision Recommendation

Contributed to a Python/FastAPI service that recommends active trial plans and permitted extension options for SPM decision workflows while maintaining an auditable human-review path.

Python FastAPI XGBoost Vector Databases

Conversational AI proof of concept

GenAI Chatbot for URL & Opportunity Creation

Led a conversational-AI proof of concept combining BlenderBot and vector retrieval for sales-rep intent recognition, with workflow UI and Search Service REST APIs for guided URL and opportunity creation.

BlenderBot JavaScript Spring Boot Vector Database

05 / Architecture & system design

Built for throughput, recovery, and safe promotion.

The platform work brings together event-driven services, data extraction, search indexing, and production controls. This conceptual view reflects the technologies and flow described in the resume—not internal implementation detail.

Oracle data foundationtables · materialized views · packages

→

Spark extractionindependent monthly & rolling loads

→

Kafka consumptionmulti-pod, parallel consumers

→

OpenSearch indexingparallel bulk workers

→

Safe alias promotionsearch, APIs & workflows

Event-driven architecture Independent extraction and sequential indexing Recovery and out-of-order-load safeguards OCI-managed Kubernetes deployment

06 / Technical expertise