Software Engineer, Kernel Development and Optimization (AI)

TL;DR

Software Engineer, Kernel Development and Optimization (C++/AI): Developing performance-critical kernels for AI compute hardware with an accent on ML and HPC workloads. Focus on designing GPU-style kernels, optimizing throughput, and implementing low-level concurrency and synchronization strategies.

Location: Hybrid based out of Warsaw or Gdansk, Poland

Company

hirify.global is building next-generation AI compute, developing high-performance RISC-V CPUs and an AI platform to revolutionize performance and cost efficiency.

What you will do

• Design, implement, and optimize GPU-style kernels such as matrix multiplication, attention primitives, and data-movement operations.
• Identify performance bottlenecks and deliver measurable throughput improvements.
• Contribute to host-side orchestration code and parallelization strategies.
• Develop micro-benchmarks, regression tests, and tooling to ensure correctness and sustained performance.
• Collaborate with compiler, runtime, ML, and hardware teams to integrate kernels into production systems.

Requirements

• Strong C++ systems engineering experience with performance-critical or low-level software.
• Proficiency in reasoning about concurrency, synchronization, latency hiding, and compute vs. memory trade-offs.
• Data-driven approach using profiling and benchmarking to guide optimization decisions.
• Ability to debug complex runtime or kernel-level issues in large codebases.
• Must be eligible to access U.S. export-controlled technology (compliance with EAR).

Culture & Benefits

• Highly competitive compensation package.
• Opportunity to work on cutting-edge AI hardware and RISC-V CPU technology.
• Learning experience in writing accelerator kernels outside traditional CUDA ecosystems.
• Practical exposure to AI-assisted and agentic workflows for kernel generation and debugging.