ProductsSiFive Core IPSiFive Core DesignerSoftwareBoardsSoC IPCustom SiliconDocumentationCustomer Support
TechnologyRISC-VSiFive IntelligenceScalable MicroarchitecturesMix+MatchRelentless InnovationTrace+DebugSecurity
Machine Learning/AI Software Engineer
SiFive is an idea-to-silicon company founded by the inventors of RISC-V to simplify the design and production of custom SoCs.
As the leading commercial provider of RISC-V processor IP, SiFive is on a mission to help engineers design custom chips for domain-specific solutions for many markets, including 5G, edge AI, enterprise networking, storage, and consumer devices.
Industry-leading innovators, including six of the top ten semiconductor companies, are working with SiFive thanks to our proven success, deep expertise, and rich partner ecosystem. With SiFive’s rich IP ecosystem and accessible design platform, every market has access to the development of workload-focused hardware needed to design next-generation products.
As new RISC-V ISA extensions are prototyped, implemented, and standardized, SiFive will remain at the forefront of RISC-V innovation by shipping the highest performance, the lowest power RISC-V processors in the world. In order to achieve this goal, we need specialists in vector extensions and heterogeneous software architecture. In the SiFive Vector team, you will build a Heterogeneous computing software stack based on RISC-V architecture and vector extension for AI/ML applications, and also the high-performance vector kernel/library solutions.
We believe that multi-functional cooperation can spark innovation and reduce production cycle time. Thus, the work environment is especially great if you are a team-player. Through collaboration with the Hardware team, we are dedicated to analyzing and improving the efficiency of our systems.
- AI/ML or Heterogeneous computing infrastructure development
- Parallel programming & optimization
- Deep learning application/middleware integration
- Open-source software development/ integration
- Co-work with SW & HW members for performance exploring and tuning
- 3+ years experience working in a large code base software, and strong c/c++ programming.
- 3+ years Software design and programming experience in C/C++/Python for development, debugging, testing and performance analysis.
- Familiarity with AI Compiler (like MLIR/TVM/Glow etc).
- Familiarity with AI Runtime (like TFRT/IREE/ONNX-runtime etc).
- Basic understanding of Deep learning
- Open source upstreaming experience is plus
- Good understanding of both CPU and System Architecture is plus
Besides the above, if you have rich experience in any heterogeneous computing runtime, you are the professional we are looking for!