Sushant
Backend Engineer • Distributed Systems & Databases
I design and implement highly performant, distributed, and resilient backend systems. Currently at Conology GmbH, I focus on building multi-tenant architectures, securing microservices, and orchestrating robust event-driven networks. My technical write-ups focus on system design, database internals, and systems post-mortems.
Core Specializations
Distributed Systems & Messaging
Custom message broker design (Mini-Kafka in Java), Raft consensus engines, and Cassandra anti-entropy database repair algorithms using Merkle Trees.
Database Engineering & Connection Pooling
PostgreSQL process-per-connection architecture, HikariCP lock-free ConcurrentBag internals, PgPool HA clustering, and DynamoDB schemas.
Concurrency & JVM High-Throughput
Reactive Java programming (Reactor/WebFlux), multithreaded batch processing (300M+ EOD records), and Redis distributed lock patterns.
Platform Infrastructure & Cloud Native
Kubernetes persistent storage layers (PV/PVC/SC), OAuth 2.0 / Keycloak multi-tenant secure gateways, and Camunda workflow engines.
Professional Journey
Conology GmbH
March 2024 — PresentBackend Software Engineer
Focused on multi-tenant core services, integration gateways, and event-driven data streaming. Maintained identity security boundaries and optimized distributed query flows across enterprise platforms.
Enfusion
Jan 2022 — March 2024Software Engineer (Backend Core)
Maintained core trading system pipelines and transactional order matching engines. Designed low-latency institutional broker gateways and optimized end-of-day batch processing routines.
> Featured Deep Dives
Detailed architectural write-ups and system implementations.
Deep Dive: Building a Distributed Message Broker from Scratch
Designing and implementing a Mini-Kafka message broker from scratch in Java, covering sequential I/O storage, consumer offset tracking, partitioning strategies, and Raft consensus.
Deep Dive: PostgreSQL Connection Pooling & HikariCP Internals in Spring Boot
A highly technical masterclass exploring PostgreSQL's process-per-connection architecture, OS socket constraints, HikariCP's lock-free ConcurrentBag internals, and precision performance tuning for Spring Boot.
Why Pgpool-II HA Mode Breaks Camunda 7 in Kubernetes: A Systems Post-Mortem
A deep-dive analysis of why deploying Pgpool-II in High Availability (Watchdog) mode via Helm charts triggers catastrophic transactional failures and split-brains in Camunda 7 workflow engines.
Demystifying Merkle Trees: Anti-Entropy in Cassandra and Blockchains
A deep dive into Merkle Tree internals, their cryptographic validation pathways, and their crucial role in managing distributed database repairs and blockchain block verifications.
> Recent Technical Notes
Shorter notes on algorithms, quick how-tos, and developer setups.