Eugene Ossipov

Owned end-to-end modernization of a mission-critical daily funding system. Audited the legacy .NET/SharePoint-based system, identified single points of failure, and produced the blueprint for a high-availability, low-latency microservices architecture. Architected and built the core DFR REST API engine on an asynchronous, event-driven framework (Java, Spring Boot, Spring Cloud), implementing CQRS and Event Sourcing with the Axon Framework to decouple heavy transactional mutations from real-time read projections. Led a hands-on squad of 4 engineers through deployment, code reviews, and high-throughput data processing standards, delivering the platform to production.

Acted as Kafka architect/technical SME for a production streaming data platform serving GoodLabs and its customers. Profiled complex Kafka Streams processing topologies, eliminated redundant re-partitioning, and tuned RocksDB state stores to maximize throughput and minimize latency. Designed an architectural refactoring plan resolving data skew and enforcing exactly-once processing semantics across distributed event-driven microservices. Engineered an automated performance benchmarking framework using Linux system metrics, JVM runtimes, Kafka broker logs, and Kubernetes cgroups to systematically measure lag, throughput, and resource bottlenecks. Authored a deterministic execution playbook that let development teams apply tuning updates without breaking schema compatibility or causing downtime.

Deconstructed a monolithic J2EE (Java 6) pension administration system, isolating domain boundaries to map out a cloud-native, event-driven architecture. Engineered a production-ready solution prototype for the Membership Enrolment system with end-to-end data pipelines in Java 11 and Kotlin. Built high-throughput event microservices integrating Kafka with Zeebe and Axon to coordinate long-running distributed workflows and sagas. Optimized read/write throughput for streaming state updates using CQRS paired with heavily index-optimized PostgreSQL datastores.

Architected and developed 5 core streaming APIs using Java 8, Kotlin, and Spring Cloud with Apache Kafka as the event backbone, processing transactional financial data across Pivotal Cloud Foundry. Engineered a mission-critical fund accounting pipeline that scaled to over 1 million real-time streaming requests per day, automating legacy manual reporting layers and saving around $100,000 CAD annually. Designed Kafka partition strategies and consumer group management to guarantee message ordering under a strict two-nines production SLA for top-tier global banking clients. Deployed automated CI/CD pipelines (Jenkins, IBM UrbanCode Deploy), enabling zero-downtime canary deployments.

Co-developed the core microservices architecture for a high-security mobile banking application using Kotlin, Kafka, and Spring Boot to consume real-time Open Banking APIs. Implemented event-driven Kafka pipelines processing account data streams with asynchronous synchronization into distributed MongoDB clusters. Built secure, low-latency integrations with Google Firebase to push real-time transaction states and ledger updates to the mobile client layer. Focused on high-availability state management and partition strategy optimization to prevent out-of-order execution in critical financial transactions.