Software Engineer II (JVM)

Overview

Software Engineer II (JVM) working on the backend systems that move enterprise data through Nexla: JVM services, the Kafka-based streaming runtime, and the distributed processing beneath it.

What you'll do

• Design, build and ship core framework features, including schema evolution, pagination, rate limiting, circuit breaking, and automated retries and self-recovery.
• Write, document and use the SDKs and interfaces other teams build on.
• Operate the streaming and processing path in production, including Kafka, Spark and Flink workloads, checkpointing, backpressure and failure recovery.
• Make design decisions within your scope and contribute to decisions above it, including tradeoffs around multi-tenancy, concurrency and latency.
• Trace failures across Kafka clusters, distributed runtimes and third-party APIs, fix them, and build guardrails to prevent recurrence.
• Take part in code review and write RFCs, runbooks and docs.

What you'll need

• 5 to 8 years of software engineering experience.
• Some of that experience on backend platforms, services or SDKs rather than only end-user applications.
• Strong Java, including JVM memory management, concurrency models and non-blocking I/O.
• Hands-on experience running distributed systems in production, such as Kafka, Spark, Flink or similar.
• Comfort with the modern data stack and protocols: Snowflake, Databricks, gRPC, GraphQL, webhooks and CDC.
• Experience owning production systems and applying good operational discipline.
• Use of AI tools in your actual work, with the ability to say specifically where they help and where they mislead.

Nice to have

• Contributions to open-source data projects such as Airbyte, Flink, Debezium or Arrow.
• Experience at an iPaaS or ELT company such as MuleSoft, Fivetran, Informatica, Airbyte or similar.
• Deep Kubernetes and cloud-native experience, including Strimzi for Kafka.
• Security architecture experience: OAuth 2.0, mTLS and encrypted credential management at scale.

Details

• Location: Bengaluru, Karnataka, India.
• Hybrid, two days per week in office.
• Overlap with morning Pacific hours is required for syncs, design reviews and collaboration with US-based teams.
• Hiring process: coding, low-level design, high-level design and a hiring manager conversation, about two weeks end to end.