Staff Compiler Performance Engineer
jobgether
Posted Oct 16, 2026
- Full-time
- Remote
- Security & IT
Job description
**This position is listed on behalf of a partner company, who manages all applications and next steps. Our partner is looking for a Staff Compiler Performance Engineer based in India.** Join a team of compiler experts working on advanced Java performance optimization for enterprise-scale applications. In this senior technical leadership role, you will shape the performance strategy and long-term architecture of a mature, production-grade just-in-time (JIT) compiler built on LLVM. You will identify opportunities for significant performance gains across benchmarks and real-world customer workloads, translating them into innovative optimization strategies. Working at the intersection of compiler technology, runtime systems, and garbage collection, you will influence critical architectural decisions and drive improvements across the software stack. You will also contribute to hardware enablement, guide complex engineering initiatives, and mentor experienced engineers within a globally distributed team. This is an opportunity to solve challenging performance problems, influence open-source compiler development, and help deliver exceptional Java application performance at enterprise scale. ### Accountabilities - Performance Strategy and Technical Roadmap: Define the compiler's performance direction by analyzing benchmarks, production workloads, and customer requirements to identify the most impactful optimization opportunities. - Compiler Optimization and Code Generation: Lead the design and implementation of complex, cross-cutting optimizations across compiler optimization pipelines and code generation systems. - Architecture and Technical Direction: Make key architectural decisions concerning the optimization pipeline, compilation-time budgets, tiered compilation strategies, and the balance between peak performance, startup time, and code size. - Runtime and Garbage Collection Integration: Drive the design of interfaces between the compiler, runtime, and garbage collector, ensuring effective coordination around speculation, deoptimization, and memory-management barriers. - Hardware and Product Enablement: Lead engineering efforts to support new hardware platforms and product capabilities, ensuring compiler performance and architectural scalability. - Technical Leadership and Mentoring: Provide technical guidance to senior engineers, review critical designs and code, and establish best practices for performance measurement, benchmarking, and validation. - Performance Analysis: Investigate performance bottlenecks at the microarchitecture level and translate findings into measurable improvements in compiler efficiency and generated machine code. - LLVM Community Engagement: Represent the engineering team in the LLVM ecosystem by contributing to upstream development, participating in architectural discussions, and advancing compiler technologies for managed programming languages. - Cross-Functional Collaboration: Partner with runtime, garbage collection, and product engineering teams to align technical decisions with broader platform objectives and long-term product requirements. ## Requirements - Professional Experience: At least 8 years of professional experience developing optimizing compilers or just-in-time (JIT) compilers. - Technical Leadership: Demonstrated experience leading the architecture and design of major compiler components or optimization frameworks deployed in production environments. - C++ Expertise: Expert-level proficiency in C++ and experience developing, maintaining, and evolving large, long-lived software codebases. - Compiler Optimization: Deep knowledge of compiler optimization techniques, code generation, and optimization pipeline architecture. - Performance Engineering: Advanced performance analysis skills, including microarchitecture-level investigation, benchmarking, profiling, and bottleneck identification. - Optimization Trade-Offs: Strong understanding of the trade-offs between peak execution performance, <