Compiler Engineer (Embedded Software & AI)

About the role

We are looking for a Compiler Engineer to design and implement embedded software libraries and low‑level runtimes that enable AI and DSP workloads on our custom hardware. The role bridges hardware and software, delivering production‑ready drivers, compiler passes and performance tooling for edge devices.

Key responsibilities

• Design and implement embedded software libraries and low‑level runtime components for MCU and accelerator targets.
• Develop and maintain MLIR/LLVM passes, code generation kernels, and the overall compiler path that maps AI/DSP primitives to hardware.
• Create and evolve a benchmarking and profiling framework with reproducible tests, dashboards and CI‑integrated regression gates.
• Strengthen build, test and continuous‑integration pipelines to ensure predictable releases and easy artifact consumption.
• Collaborate with hardware, architecture and customer‑facing teams, write precise specifications and documentation, and translate feedback into roadmap items.

Required profile

• 5+ years of experience building low‑level software or compilers and shipping production code.
• Strong C++ and Python programming skills.
• Hands‑on experience with embedded systems and compiler design, including memory and concurrency fundamentals.
• Ability to read hardware datasheets and work effectively at the HW/SW boundary.
• Proven track record of performance profiling, tracing and optimization on embedded or accelerator targets.
• Clear writing, good documentation habits and a collaborative mindset.

Required skills

• C++
• Python
• MLIR
• LLVM
• Embedded systems development
• Hardware‑software integration
• Performance profiling and optimization
• CI/CD pipelines (e.g., GitHub Actions)
• TensorFlow Lite for Micro, TVM, IREE (nice to have)
• DirectML, OpenCL, CUDA (nice to have)
• Quantization and fixed‑point arithmetic (nice to have)