Bhavesh Nakum
Senior Software Engineer · Backend, Distributed Systems & Big Data · Cloud & AI-Augmented Engineering
Bengaluru, India bhaveshnakumcoed@gmail.com linkedin.com/in/bhaveshnakum7
This Gmail address is for professional enquiries only. It is the same one on the PDF résumé. For anything else, use Contact me in the footer.
Summary
Senior Software Engineer with 8+ years of experience designing, developing, and operating high-performance backend systems, distributed microservices, and large-scale data platforms in Java and Python. Deep expertise across event-driven architecture (Apache Kafka, Apache Flink), big data processing (Apache Spark, Apache Airflow, Hive, HDFS), and multi-tenant, multi-region SaaS platform design with a strong focus on high availability, fault tolerance, and secure API development (OAuth, OWASP). Hands-on experience designing AI agent-driven workflows (including an agentic system that generates data schemas from natural-language chat input and a self-service, automated data consumption layer), alongside daily use of AI-augmented development tools (Cursor, Claude Code, GitHub Copilot, Qodo Merge) across coding, design, PR review, and testing. Skilled in cloud-native engineering across Microsoft Azure, Google Cloud Platform, and AWS, with hands-on experience in Kubernetes, Docker, and CI/CD pipeline automation. Proven technical leader and lead developer with a track record of architecting production systems end-to-end, mentoring engineering teams, participating in technical hiring, and collaborating cross-functionally with product, data science, and business stakeholders across e-commerce, healthcare, and enterprise IT environments.
Experience
Senior Software Engineer, eBay
Owns eBay's real-time data pipelines, plus the team's AI schema generator and self-service data layer.
- Data Platform Ownership: Member of eBay's core Data team, owning end-to-end design, development, and maintenance of real-time and batch data pipelines across the eBay platform; responsible for reliable delivery of eBay Live feature data to downstream data consumers using Apache Flink, Apache Kafka, SQL, and Hadoop.
- AI Agent-Driven Schema Onboarding: Designed and implemented an AI agent workflow within the team's data consumption layer that automatically generates data schemas from natural-language chat input and user-provided data, replacing manual schema-definition forms and streamlining the onboarding of new datasets.
- Self-Service Automated Data Consumption: Built an automated data consumption layer enabling downstream consumers to write and run their own SQL queries to extract data on a batch, scheduled, or on-demand basis, eliminating the need for the data management platform team to manually author extraction queries and pipelines for each consumer request.
- AI-Augmented Development: Use Cursor and Claude Code for coding, PR reviews, design, and testing as part of daily engineering practice.
Senior Software Engineer (Lead), Walmart
Led 6 engineers building Kafka-based pricing microservices across multiple countries.
- Distributed, Event-Driven Architecture: Architected Java and Spring Boot microservices on a Kafka-based event-driven architecture, processing 5K+ events per second with sub-second latency to power dynamic repricing across millions of SKUs.
- Multi-Tenant SaaS Platform Design: Designed the schema and key strategy for a multi-market pricing platform spanning Canada, Mexico, Chile, and additional markets, with configurable tenant-level isolation supporting both database-level separation and tenant-based table partitioning to meet country-specific data compliance requirements.
- High Availability Architecture: Deployed database and application infrastructure across 2 active regions and 1 replica region to eliminate single-region outage risk and optimize regional availability and latency.
- CI/CD, Containerization & Deployment: Built Docker and Kubernetes-based deployment templates (including Helm) within Walmart's internal release pipeline, packaging code into deployable artifacts and applying blue-green deployment strategy for Kafka-based services and canary deployment for API-based applications, with automated test execution, post-deployment code coverage checks, metrics exposure, and autoscaling-based alerting.
- Test-Driven Development & Code Quality: Practiced Test-Driven Development, writing tests before implementation with automated coverage via JUnit, Mockito, and Cucumber; led a structured PR review and merge process for a team of 6 engineers, enforcing design doc adherence, the KISS principle, and a 60/40 positive-to-negative test case ratio.
- Secure, Well-Architected APIs: Designed, deployed, and maintained RESTful APIs and Backend-for-Frontend (BFF) services applying OAuth-based authentication and OWASP secure coding practices, tailored to different client application needs.
- Observability & Cost Optimization: Designed and delivered a reusable, plug-and-play Micrometer metrics module adopted across engineering teams for self-service observability and custom Grafana alerting; reduced infrastructure cost and improved availability by fine-tuning CPU/memory allocation, enabling autoscaling, and consolidating deployables.
- AI-Augmented Development: Used GitHub Copilot for development work and design assistance, and Qodo Merge for automated PR review on GitHub, integrated into daily engineering workflow.
- Technical Leadership & Collaboration: Led and mentored a cross-functional team of 6 engineers, participated in hiring and interview panels for vendor and FTE roles, and partnered quarterly with product managers and data scientists to define feature roadmaps and plan cross-functional releases.
Software Developer, Walmart
Built repricing and anomaly-detection pipelines that pulled over 1M mispriced items off the Walmart catalog.
- Pipeline Engineering: Built real-time anomaly detection and repricing pipelines, removing 1M+ mispriced items from the Walmart catalog.
- Seller Tools: Developed pricing automation tools for sellers, reducing manual overhead and improving competitiveness.
- Reliability: Enhanced system reliability with full-stack alerting and dashboarding for production-grade observability.
Data Engineer, UnitedHealth Group
Built HIPAA-compliant ETL pipelines on Azure Data Factory and Databricks for healthcare data.
- ETL Design: Designed scalable ETL pipelines using Azure (Data Factory, Databricks, Synapse) to ingest and transform healthcare data across supported asset classes and data domains.
- Automation: Developed PySpark-based transformation logic and modular Airflow DAGs to accelerate pipeline development while ensuring HIPAA compliance.
Systems Engineer, Tata Consultancy Services
Migrated legacy data warehouses to Azure and cut query costs 25%.
- Cloud Migration: Migrated legacy data warehousing platforms to a Hadoop-based Azure ecosystem for Proximus (Belgium).
- Optimization: Rewrote HQL queries to optimize performance and reduce compute costs by 25%.
- Modernization Tools: Engineered a Python-based tool to convert Teradata SQL to Hive and PySpark scripts using Python and Regex for Bank of America.
Projects
Real-Time Emotion-to-Emoji Detector
Built a browser tool that reads your facial expression through the webcam and shows a matching emoji.
- Live capture: Captures your face live through the webcam, right in the browser, no app or upload required.
- Expression detection: Reads the expression and picks a matching emoji: happy, sad, angry, surprised, fearful, disgusted, or neutral.
-
Turn your camera on and it'll guess how you're feeling.
Runs entirely in your browser. Nothing leaves your device.
Additional highlights
- Agentic AI System Design: Hands-on experience designing and shipping an AI agent-driven workflow for automated data schema generation and a self-service, automated data consumption layer, moving a previously manual, team-dependent process to a fully automated, consumer-driven path.
- Mentorship & Hiring: Mentored new hires and junior engineers; participated in hiring and interview panels for vendor and FTE roles.
- AI-Augmented Development: Practical, day-to-day use of AI coding tools (Cursor, Claude Code, GitHub Copilot, Qodo Merge) and general-purpose LLM tools (Claude, ChatGPT) to accelerate development, elevate code review quality, and support research and technical communication.
- Legacy & Proprietary System Modernization: Consistent track record diving into proprietary and legacy systems (internal release pipelines, Teradata data warehouses, Hadoop platforms) to understand their inner workings and strategically modernize them.
- Regulated & Enterprise Environments: Experience delivering technology solutions in compliance-driven settings (HIPAA-regulated healthcare data) and for large enterprise clients (Proximus/Belgium, Bank of America).
- CI/CD: Automated CI/CD pipelines by integrating functional, integration, end-to-end, and performance testing into the deployment process.
- Engineering Standards: Passionate about performance optimization, code quality, testability, and technical documentation for high-concurrency, high-throughput production systems.
Skills
- Languages
- JavaPythonSQL
- Backend Development & APIs
- Spring BootRESTful Web ServicesMicroservicesBackend-for-Frontend (BFF)Distributed SystemsEvent-Driven ArchitectureJava Concurrency (Threads, Executors, CompletableFuture, Reactive Patterns)
- Security
- OAuthOWASP Secure Coding Practices
- AI Agents & Automation
- Agentic Workflow DesignAutomated Schema Generation from Natural LanguageSelf-Service Data Pipeline Automation
- Big Data & Streaming
- Apache KafkaApache SparkApache FlinkApache AirflowHiveHDFSSqoop
- Data Modeling & Architecture
- Schema DesignKey Strategy & PartitioningGrain DefinitionMulti-Tenant Data IsolationNoSQL Data Modeling
- Cloud Platforms
- Microsoft Azure (Data Factory, Databricks, Synapse, Key Vault)Google Cloud Platform (DataProc, BigQuery, GCS)AWS (Lambda, S3, SQS)
- High Availability & Resilience
- Multi-Region ReplicationHigh AvailabilityFault ToleranceBlue-Green & Canary DeploymentsAutoscaling
- Databases
- PostgreSQLMySQLAzure Cosmos DBCassandra (Relational & NoSQL)
- CI/CD & DevOps
- JenkinsDockerKubernetesHelmMavenGitCI/CD Pipeline AutomationInfrastructure AutomationAgile/Scrum/SDLC
- Testing & Quality
- JUnitMockitoCucumberTest-Driven Development (TDD)Unit/Functional/E2E Testing
- Observability & Monitoring
- GrafanaMicrometerCustom Metrics & AlertingMonitoring & Dashboarding
- AI-Augmented Engineering
- CursorClaude CodeGitHub CopilotQodo MergeClaudeChatGPT
- Familiarity (exposure, non-production depth)
- SnowflakeGraphQLHibernate/JPAElasticsearch
Education & certifications
- B.Tech, Computer Engineering, SVNIT, Surat (2018) · CGPA 7.18/10
- Microsoft Certified: Azure Fundamentals
- Apache Airflow Fundamentals
Frequently asked questions
Who is Bhavesh Nakum?
Bhavesh Nakum is a Senior Software Engineer based in Bengaluru, India, currently working on eBay's core Data team.
What does Bhavesh Nakum do?
He designs and builds backend systems and distributed data pipelines, with hands-on work in event-driven architecture (Apache Kafka, Apache Flink), big data processing, and AI agent-driven workflows.
Where has Bhavesh Nakum worked?
eBay, Walmart (as Senior Software Engineer, Lead), UnitedHealth Group, and Tata Consultancy Services, spanning backend engineering, data engineering, and cloud migration work since 2018.
Is Bhavesh Nakum a lead developer?
Yes. At Walmart he led and mentored a cross-functional team of 6 engineers as Senior Software Engineer (Lead), and continues to take on technical leadership and mentoring responsibilities.
What technologies does Bhavesh Nakum work with?
Java, Python, Spring Boot, Apache Kafka, Apache Flink, Apache Spark, Kubernetes, Docker, and cloud platforms including Microsoft Azure, Google Cloud Platform, and AWS.