Role summaryYou’ll own the data replication layer between heterogeneous database systems — designing the node topology, configuring capture and routing, and keeping bi-directional synchronisation running in production. This is a hands-on engineering role with a strong operational component: when replication stalls at 2am, you’re the person who works out why. SymmetricDS is the platform, but the job is replication engineering. You’ll need to be as comfortable reading an execution plan and a transaction log as you are configuring a router. What you’ll be doingTopology and configuration
• Design node groups and group links, and decide push versus pull per link based on network topology and who can initiate connections
• Configure channels to isolate throughput and control ordering between unrelated data sets
• Configure triggers and routers — default, column-match, lookup-table, subselect and scripted — so each table’s changes reach only the nodes that need them
• Build table and column transforms where source and target schemas differ
• Manage node registration and the registration server Bi-directional replication
• Design conflict detection and resolution strategies per channel, and understand what each one does to data you care about
• Prevent replication loops, including on DDL changes
• Reason about ordering and causality across multi-master topologies Initial loads and schema change
• Plan and execute initial loads and selective reloads at production volume, without taking the source system down
• Handle schema evolution across nodes — re-syncing triggers after DDL, and sequencing changes so replication doesn’t break mid-deployment Running it in production
• Diagnose and clear batches stuck in error, and understand what caused the batch to fail rather than just retrying it
• Handle data gaps, tune the purge job, and keep change-capture table growth under control
• Monitor replication lag, routing backlog and throughput,
and alert on them before users notice
• Capacity planning and JVM tuning, particularly for remote or bandwidth-constrained nodes
• Upgrades, patching and version compatibility across nodes What we’re looking forExperience — 5+ years in data replication, database integration or data engineering, including hands-on production work with SymmetricDS. Time spent supporting a live replication estate counts for more with us than implementation projects alone. SQL and database depth — solid SQL across at least two of PostgreSQL, MySQL/MariaDB, Oracle or SQL Server. You should be comfortable with execution plans, indexing, locking and transaction isolation, because replication problems are usually database problems wearing a disguise. Change data capture — you understand how trigger-based capture works and what it costs the source system, how log-based capture differs, and when each is the right choice. Java — routers, transforms and extension points are Java and BeanShell. You don’t need to be a full-time Java developer, but you need to read and write it confidently. Linux and the JVM — service management, log analysis, heap and GC tuning. Operational temperament — you can work an incident through an unfamiliar system from symptom to root cause, and you document what you found. Nice to have
• SymmetricDS Pro — the web console, clustering, bulk loading, data validation, log-based capture
• Other CDC and replication tools: Debezium, Oracle GoldenGate, AWS DMS, Qlik Replicate, Syniti Replicate
• Streaming or warehouse targets — Kafka, Snowflake, BigQuery, Redshift, Databricks
• File synchronisation across nodes
• Docker and Kubernetes
• Multi-site, retail/POS or occasionally-connected deployments
• Experience with an eCommerce or ERP data model How we selectA technical conversation built on real scenarios — a batch sitting in error, a data gap you need to explain, a conflict on a bi-directional channel, an initial load that has to run against a live system. We’re interested in how you work the problem.
📌 Symmetric DS (India)
🏢 Codem
📍 India