If your jam is low-level programming, high-performance DSPs, and working close to the metal, this one’s for you. What You Will Do: Profile, analyze, and optimize performance, power, and area trade-offs in software designs; Develop and optimize embedded software for high-performance multicore DSP platforms; Leverage SIMD, VLIW, and fractional arithmetic techniques to deliver efficient signal processing and linear algebra algorithms; Work closely with hardware design teams to understand instruction set architectures (ISA) and micro-architectures details, ensuring software takes full advantage of the hardware; Integrate DMA strategies to maximize data throughput; Build, use, and maintain pre-silicon validation platforms such such as virtual prototypes for early software development and testing. What You Will Bring: Strong experience programming for embedded processors or low-level systems; Solid understanding of computer architecture and micro-architecture fundamentals; Hands-on experience with SIMD and VLIW programming models; Knowledge of DMA integration in embedded systems; Familiarity with fractional arithmetic in DSP applications; Experience working with instruction set architecture (ISA) documentation or development; Exposure to virtual prototypes or pre-silicon validation platforms; Proficiency in C/C++, Assembly, Intrinsics; Ability to analyze performance bottlenecks and optimize for PPA. Bonus Points: Individuals who have worked with CEVA DSPs, particularly the SensePro family (SP500/SP1000) and MRD accelerators, will be exceptionally well-suited for this role; Familiarity with CEVA’s instruction set, toolchains, and performance optimization techniques will be highly advantageous, especially in the context of high-performance signal processing for sensing applications. Location: Boston MA (hybrid).