Senior Compiler Engineer

You will be part of a high-performance systems team focused on optimising code generation and execution for a next-generation compute architecture. This role involves working deeply across compiler development and hardware interaction to improve performance and enable new architectural capabilities.

You will be responsible for:

• Enhancing and contributing to the development of an LLVM-based compiler
• Collaborating on the programming model for a high-performance computing platform
• Working closely with hardware design teams to co-optimise compiler strategies with architecture
• Building and optimising custom front-end and back-end compiler passes
• Supporting internal and external stakeholders on compiler-related challenges

Ideal Candidate

• You have an MS or PhD in Computer Science with 5+ years of experience in compiler development
• You have strong proficiency in C++ (C++11 or later)
• You have hands-on experience with LLVM or GCC codebases
• You have a solid foundation in graph algorithms and data structures
• You have strong knowledge of front-end and back-end compiler techniques
• You have experience working on performance optimisation at the systems or compiler level

Nice to Have

• You have experience with compiler strategies for parallel architectures such as GPUs or DSPs
• You have knowledge of end-to-end toolchains including compilers, linkers, and debuggers

What’s on Offer

• Competitive compensation with meaningful equity
• High-impact role in a technically strong, low-bureaucracy environment
• Opportunity to build long-term career and work on cutting-edge systems

About the employer

Our client is a Silicon Valley–based deep-tech company building a new compute architecture for real-time AI at the edge. Founded by engineers from leading research backgrounds, the focus is on solving the gaps in current neural processing approaches through tight integration of hardware and software.

The platform is built to run both neural network inference and conventional compute workloads efficiently across a wide range of edge devices. Unlike typical accelerators that only handle parts of an ML graph, this architecture supports end-to-end execution, including both neural network graph code and standard C++ DSP and control code, enabling greater flexibility and performance in real-world deployments.