Our company

We are Backgammon Galaxy, the biggest backgammon world site in the world. It's a social game where you play for rating points.

Project description

We are looking for a Systems/performance engineer for optimization and scalability.

• C/C++ performance
• SIMD intrinsics (AVX2/FMA)
• profiling (perf/VTune)
• numeric kernels
• cache/memory optimization
• PGO/LTO
• multithreading.

For this project (profiling + SIMD kernels + possible quantization + multithreading) we are looking for a systems/performance engineer, not a general application developer. An optimal “stack” is:

• Languages: C (primarily) and C++ (optional, for test harnesses/bench tooling), plus a small amount of Python for benchmarking automation.
• CPU optimization: x86-64 SIMD intrinsics (SSE2/AVX2/FMA; optional AVX-512) via <immintrin.h>; runtime CPU feature dispatch.
• Compilers/toolchains: Clang/LLVM and GCC (both), using -O3, LTO, and PGO.
• Profiling: Linux perf (+ flamegraph), Valgrind/callgrind; optional Intel VTune if on Intel hardware.
• Correctness/perf testing: GoogleTest/Catch2 (or plain C harness), deterministic regression suite; CI on GitHub Actions (Linux) with benchmark baselines.
• Parallelism: POSIX threads / OpenMP (pick one; OpenMP is faster to implement, pthreads gives tighter control).
• Build system: CMake (or Meson) plus Ninja; keep Autotools compatibility only if required by upstream.

Project duration

• 3-6 months.

Qualifications

• Senior developer with loads of experience.

Salary budget

• The salary will be milestone and/or project based, approx. $25,000.