Stateful actors on Apache Flink 2.x and Java 21 — durable per-key state, exactly-once messaging, Kafka and Kinesis I/O, Kubernetes-native deployment. Continues the Apache Stateful Functions programming model on the modern Flink line.

Riffl - generic streaming data ingestion framework

Wrappers for Apache Kafka java clients

3 Problems from SP04 to be solved using HashMap/ HashSet instead of TreeMap/ TreeSet.

Exactly-once Kafka delayed-message relay: re-deliver each record to a destination topic at the time its producer asked for, observed exactly-once by read_committed consumers.

Kafka-shaped Message Broker built in Java

Kafka→Solace event bridge with at-least-once delivery, per-key ordering, and Redis-backed idempotency. Java 21 · Spring Boot 3.5 · raw JCSMP.

Exactly-Once Processing Across Kafka and Databases: Using Kafka Transactions + Idempotent Writes

Java + Kafka stream processor with Postgres lookup enrichment, exactly-once semantics, and broker-restart-verified idempotency

Queue benchmark harness: one fan-out workload, one messaging contract, many backends.

Distributed event processing with Apache Spark Structured Streaming: Kafka ingestion, stateful deduplication, watermark-based window aggregation, and reproducible Parquet/Kafka outputs.

A production-grade real-time streaming pipeline ingesting multi-source events (REST, IoT, WebSocket) into Apache Kafka (Avro & Schema Registry), statefully processing streams via Apache Flink with exactly-once Delta Lake storage, dbt transformations, and Grafana dashboards.

Real-time usage metering & monetization platform — Kafka pipeline, double-entry ledger, PostgreSQL RLS multi-tenancy, WebFlux vs virtual-threads benchmark

Exactly-Once Processing Across Kafka and Databases: Using the Outbox Pattern

Kafka exactly-once payments backend with Outbox Pattern, Debezium CDC, and idempotency tracking.