senior backend engineer

Prakash Verma

Senior Java Backend Developer

8.5 years building event-driven, cloud-native backend systems for telecom, mobility, insurance and payments. I design APIs meant to survive contact with production, and systems built to be understood by the next engineer, not just this one.

||||

8.5yr

experience

4

companies

4

domains shipped

FIG. 01 — SYSTEM ARCHITECTURE
AWS CLOUD · EKSEVENT BUS · KAFKATELECOM DOMAIN DATAClientAPI GatewayKeycloakOrder SvcInventoryTemporalPostgreSQLMongoDB

Hover any component — or trace a request through the system.

PATTERN: EVENT-DRIVEN MICROSERVICESSTATUS: PRODUCTION-GRADE
Java 17/Spring Boot/Microservices/Apache Kafka/REST API Design/Kubernetes/System Design/PostgreSQL/MongoDB/OpenAPI / Swagger/Clean Code/Event-Driven Architecture/Java 17/Spring Boot/Microservices/Apache Kafka/REST API Design/Kubernetes/System Design/PostgreSQL/MongoDB/OpenAPI / Swagger/Clean Code/Event-Driven Architecture/

specification

I design for the day everything goes wrong at once.

Senior Java backend engineer specializing in distributed systems, microservices, and event-driven architecture — building cloud-native APIs and platforms engineered for scale, reliability, and real production behavior.

I'm a senior Java backend engineer who builds distributed systems, not just endpoints — microservices, event-driven pipelines, and cloud-native platforms designed to hold up under real production traffic, not just green tests.

Currently building O-RAN compliant orchestration software — the layer that provisions and manages telecom cloud infrastructure at scale. Before that: connected-car telemetry, medical claims adjudication, and real-time fraud detection — four domains, one consistent focus on correctness under load.

System design, to me, isn't a whiteboard exercise — it's API contracts that don't break consumers, data models that scale past the first million rows, and services that degrade gracefully instead of falling over.

Distributed SystemsMicroservicesAPI DesignEvent-Driven ArchitectureCloud-Native SystemsScalabilityReliabilitySystem Design

currently

Senior Software Consultant

T-Systems · O-RAN telecom orchestration

based in

Pune, Maharashtra, India

domains worked

Telecom · Mobility · Insurance · Payments

how I think about production systems

P.01RESILIENCE

Assume the network fails, the downstream service times out, and the queue backs up. Design for graceful degradation, not the best-case path.

P.02SCALABILITY

Stateless where possible, horizontally scalable by default — capacity is a design decision made early, not a fix applied under load.

P.03OBSERVABILITY

If it's not logged, traced, and metriced, it isn't production-ready — visibility is part of the feature, not an afterthought.

P.04MAINTAINABILITY

Code optimized for the engineer who inherits it — including me, eighteen months and one team rotation from now.

P.05API_CONTRACT

Versioned, documented, and boring by design — consumers should never be surprised by a change they didn't ask for.

system composition

Built in Layers, Not a List

Every system I've built comes down to five layers working together — from how a request comes in, to how it's stored and replayed. Expand a layer to see what actually lives there.

lang.runtime

UP

Core runtime — tracking current LTS while staying fluent across the versions still running in production. Optimized for readability first, performance second, because most incidents trace back to code nobody could safely change.

Java 8 / 11 / 17 / 25Spring BootSpring BatchHibernateJPA

api.interface

UP

How services actually talk to each other and to clients — synchronous, streaming, and bidirectional, chosen per use case rather than by default.

REST APIsgRPCWebSocketOpenAPI / SwaggerJSONXML

domain.logic

UP

Where the actual business value lives — the rules, validations, and orchestration that turn a request into a correct outcome, not just a 200 response.

Business Rule EnginesDomain-Driven DesignValidation & OrchestrationComplex Pricing / Adjudication Logic

data.eventing

UP

Persistence and the asynchronous backbone together — the right store for the access pattern, and an event/workflow layer that keeps services decoupled and durable under load.

PostgreSQLMongoDBMySQLApache KafkaTemporal

platform.cloud

UP

Infrastructure treated as a first-class concern — containers, orchestration, and pipelines that make deployment routine instead of an event.

AWSDockerKubernetesJenkinsCI/CD

quality.tooling

STABLE

The guardrails that keep 8.5 years of code honest — automated tests, static analysis, and disciplined version control.

JUnitMockitoSonarQubeGitMavenPostmanIntelliJSTS

practice.arch

STABLE

Architectural judgment — knowing when a monolith is the right call, when it isn't, and how to keep a system resilient when a dependency inevitably fails.

MicroservicesDesign PatternsResilience Patterns (Circuit Breaker, Retry)Agile (Scrum/Kanban)Clean Code

work in progress

Currently building

The two tracks I'm actively investing in outside day-to-day delivery — cloud depth, and finding where applied AI genuinely helps a backend workflow.

Cloud — AWS

IN PROGRESS

Going deeper on cloud infrastructure beyond day-to-day Kubernetes/OpenShift work.

PROGRESS68%
  • Hands-on with ECS/EKS, IAM least-privilege design, and cost-aware infrastructure decisions.
  • Working through AWS Solutions Architect – Associate material alongside production work.
  • Migrating a personal project's deployment pipeline to fully IaC with Terraform.

Applied AI for Backend Engineering

IN PROGRESS

Exploring where GenAI actually earns its place in a production backend workflow.

PROGRESS54%
  • Evaluating LLM-assisted code review and test generation inside real Spring Boot codebases.
  • Prototyping a RAG pipeline over internal API docs to speed up onboarding.
  • Studying where event streams and inference pipelines intersect — anomaly detection on Kafka topics.

experience

Track Record

8.5 years building backend systems across telecom, mobility, insurance, and payments — each role adding a harder scaling problem, and a bigger blast radius if something broke.

Leading backend design for O-RAN telecom orchestration and mobility telemetry — systems where correctness and uptime are non-negotiable.

Telecom Network Automation Platform (TNAP) — SMO / O2IMS

  • Built O-RAN compliant Service Management and Orchestration (SMO) software for telecom network automation and cloud infrastructure management.
  • Designed Spring Boot microservices for O2IMS: infrastructure inventory, resource pools, provisioning workflows, and cloud resource lifecycle.
  • Implemented REST APIs and integrated O-RAN O2 interfaces connecting SMO to O-Cloud infrastructure.
  • Built event-driven flows on Apache Kafka for resource orchestration and async inter-service communication.
  • Deployed cloud-native services on Kubernetes/OpenShift and contributed to CI/CD pipelines.
Java 17Spring BootKafkaPostgreSQLMongoDBKubernetesOpenAPIJUnit 5Mockito

Connected Car Platform — Mobility Domain

  • Contributed to a centralized platform aggregating vehicle telemetry and distributing it to brands and consumers.
  • Built and maintained microservices with Quarkus and Spring Boot in a cloud-native architecture.
  • Integrated MongoDB for real-time vehicle telemetry access; resolved SonarQube findings and wrote JUnit coverage.
  • Handled async inter-service messaging with Kafka; containerized with Docker and orchestrated on Kubernetes.
JavaSpring BootQuarkusKafkaMongoDBDockerKubernetesAWS

delivered systems

Systems I've Engineered

Most of this work sits behind client and enterprise repos, so what's below is the shape of each system rather than a link to the code.

telecom · orchestration

TNAP — SMO / O2IMS

O-RAN compliant orchestration software provisioning and managing telecom cloud infrastructure across O-Cloud environments.

Java 17Spring BootKafkaKubernetesOpenAPI

mobility · telemetry

Connected Car Platform

Centralized platform aggregating real-time vehicle telemetry and distributing it to brands and consumers.

QuarkusSpring BootKafkaMongoDBAWS

insurance · claims

CMdS Claims Engine

Real-time adjudication of medical claims for U.S. state healthcare programs, with 1,300+ configurable rules.

JavaSpring BootJPAPostgreSQL

payments · fraud

Pay360 Fraud Management

Fraud detection across 160+ transaction data points, tuned to improve both catch rate and customer acceptance.

JavaSpring BootJPAMySQL

engineering profile

Let's talk

Open to senior backend and platform roles. Email is the fastest way to reach me.

ENGINEER

Prakash Verma

ROLE

Senior Java Backend Developer

EXPERIENCE

8.5 Years

SPECIALIZATION

Distributed Systems