SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Azul is seeking a Staff Compiler Performance Engineer to lead the technical direction of Falcon, the optimizing JIT compiler in Azul Platform Prime. Falcon compiles hot Java bytecode into highly optimized machine code and is built on LLVM, currently running in production at large enterprises.
In this role, you will be the senior technical voice on a distributed team of compiler experts and a key partner to runtime, GC, and product teams. You will analyze performance of Java benchmarks and customer workloads to identify gaps in generated code and trace them to their root cause in the compiler. You will implement and improve optimizations in Falcon's middle end and back end, including inlining, loop optimizations, vectorization, redundancy elimination, and speculative optimizations and their interaction with the runtime. You will also improve code generation for x86_64 and AArch64 architectures, write tests and benchmarks to prove improvements and protect against regressions, and contribute changes upstream to LLVM where appropriate.
This is a mature compiler role focused on performance optimization, not compiler front-end development or building from scratch. You will decide where the next significant performance gains come from and lead the design of optimizations that deliver them, ensuring the compiler architecture supports the product roadmap.
Azul is the only company focused 100% on Java, reimagining how customers use Java by providing tools and insights to deliver faster, more performant, more dependable, and less expensive Java services. You will work alongside top experts worldwide contributing to the Java ecosystem.
REQUIREMENTS:
- Minimum 2+ years of professional experience in compilers, JIT compilers, or low-level performance engineering
- Strong C++ design and coding skills
- Solid understanding of compiler optimization theory: SSA, data flow analysis, loop transformations
- Ability to read assembly and reason about instruction sequence performance
NICE TO HAVE:
- Experience with LLVM IR and LLVM optimization passes, or other optimizing compilers (HotSpot C2, Graal, GCC, V8)
- Experience with profilers (perf, VTune, async profiler)
- Knowledge of x86_64 and/or AArch64 microarchitecture
- Experience with Java, Scala, Kotlin, or JVM internals
- MSc or PhD in Computer Science with compiler or performance focus
