Senior Low Latency OEMS Engineer (Java)

Jain Global • New York, NY

6d • Onsite

About The Position

We are seeking a Senior Low Latency Java Engineer to join the team that builds and runs our in-house Order and Execution Management System (OEMS). The platform is a clustered, deterministic, event-driven system built on Aeron and SBE with FIX connectivity to our executing brokers, and it is the path every order takes from trader to market. It serves trading desks in New York, London and Tokyo, trades equities today and is expanding to additional asset classes. You will own latency-critical components end to end: from the sequenced core, through the gateways, to tuning in production. We are an AI-first engineering team. Agentic coding tools are part of every engineer’s daily workflow, from design and implementation through code review to production analysis, and we invest in the rules, skills and automation that make them effective and safe on a latency-critical codebase. You should be excited to work this way and to help push it further.

Requirements

• Expert Java on a current JDK, with deep low latency experience: allocation-free coding, mechanical sympathy, cache-aware data structures, single-writer and lock-free designs.
• Hands-on experience with Aeron (transport, archive, cluster) and Agrona, or an equivalent sequencer/messaging framework (e.g. Chronicle, LMAX Disruptor).
• Binary encoding with SBE and/or Protobuf; FIX protocol at session and application level, ideally with Artio or QuickFIX/J.
• Solid understanding of sequencer and replicated-log architectures, consensus (Raft), determinism, snapshotting and replay.
• GC algorithms and tuning for low latency (ZGC, Shenandoah, G1), JIT behaviour and escape analysis, profiling with JFR or async-profiler.
• Linux performance engineering: CPU isolation and pinning, IRQ affinity, NIC and kernel tuning, tuned profiles, huge pages.
• Proficient with AI coding agents (e.g. Claude Code, Cursor, Codex): decomposing work into agent-sized tasks, writing effective context and guardrails, verifying agent output rigorously, and encoding engineering standards as machine-readable guidance.
• Working knowledge of Java NIO and zero-copy I/O; kernel-bypass networking is a plus.
• Comfortable with relational persistence (PostgreSQL) and event streaming (Kafka).
• Bachelor’s or Master’s degree in Computer Science, Engineering, or a related field.
• 6+ years of Java development, including 3+ years on low latency trading systems (OMS/EMS, exchange or broker gateways, market data, or HFT).
• Track record of owning a trading system in production on Linux, including incident response and performance investigation.
• Demonstrated use of AI coding agents on real production work, with the judgement to know when to trust, verify or override them.
• Experience with CI/CD (GitHub Actions or similar), containers (Docker/Podman), configuration management (Ansible or similar), and AWS.
• Excellent problem-solving and analytical skills.
• Strong communication skills and ability to work collaboratively within a team.
• Curiosity and rigour about how AI is changing engineering practice; comfortable being both the operator of coding agents and the accountable reviewer of their output.
• Ability to thrive in a fast-paced, high-pressure environment.

Nice To Haves

• TypeScript/React for internal tooling
• Python for automation and agent hooks
• Experience building agent skills
• MCP integrations or LLM-based developer tooling

Responsibilities

• Design and implement zero-allocation, latency-critical components in the sequenced core of the OEMS: order lifecycle, pre-trade risk checks, broker routing and position marking.
• Build and operate market-facing gateways: FIX broker connectivity (initiator and acceptor), a client-facing Aeron/SBE API, and Kafka/Protobuf integrations.
• Apply Java low latency practices on the hot path: object and buffer reuse, primitive collections, flyweight codecs, off-heap buffers, GC-free steady state.
• Measure and improve latency end to end with JMH microbenchmarks, hop-to-hop latency from production event timestamps, and distributed tracing.
• Own production performance on Linux: CPU pinning and isolation, kernel and network tuning, JVM flags and GC selection.
• Build for correctness with behaviour-driven acceptance tests, snapshot/replay-equivalence tests and integration tests.
• Work AI-first: use coding agents daily for implementation, review and investigation, and evolve the shared agent configuration (rules, skills, hooks, automated checks) that keeps AI-generated code within our low latency and correctness standards.
• Contribute to deployment and operations: containerised services, configuration-managed deployments, CI pipelines, and cross-region disaster recovery.
• Collaborate with trading, quant, DevOps and infrastructure teams.

Benefits

• Competitive compensation and potential for growth within the firm.
• A supportive and inclusive workplace culture that values diversity and collaboration.
• base salary
• discretionary bonus
• comprehensive benefits

What This Job Offers

• Job Type: Full-time
• Career Level: Senior