Job Description
About Lyte
Lyte builds perception systems for Physical AI operating in complex environments. The company combines custom silicon, integrated sensors, and software into a unified platform for robotics, mobility, and next-generation automation. Lyte is headquartered in Sunnyvale, California, with a global presence. For more information, visit www.lyte.ai
If you’re excited about building impactful technology in a dynamic, hands-on environment, we’d love to hear from you!
About the role
- We are seeking an experienced Senior DSP Engineer to join our team and help build high-
performance real-time signal processing software across a wide range of hardware architectures.
In this role, you will design and implement DSP algorithms and full embedded applications that
run on systems ranging from ARM-based platforms to VLIW processors and dedicated DSP and
vector processors. As a DSP engineer, you will work closely with the Algorithm team, as well as
system, silicon, and software teams, to translate high-level signal processing concepts into
production-grade implementations optimized for real-time embedded environments.
What you'll do
- Design and implement DSP algorithms across multiple chip architectures (ARM, VLIW,
SIMD/vector DSPs, and proprietary accelerators). - Work in tight collaboration with the Algorithm team to refine algorithm definitions,
validate numerical behavior, and support floating-point to fixed-point transitions. - Develop and optimize performance-critical kernels using chip-specific intrinsics, vector
operations, and low-level optimization techniques. - Analyze compiler output and optimize generated code through assembly review,
compiler reports, and profiling-driven tuning. - Build complete DSP processing pipelines and applications in multi-processor / multi-core
environments for real-time embedded systems. - Optimize system-level performance including compute scheduling, memory/bandwidth
usage, latency budgets, and real-time constraints. - Debug and root-cause issues related to performance, numerical stability, and real-time
correctness across hardware and software boundaries. - Collaborate cross-functionally with Algorithm, Architecture, Silicon, Verification, and
Embedded SW teams to ensure correct integration and best system-level performance.
Required Qualifications
- Bachelor’s degree in Computer Science, Electrical Engineering, or a related field
- At least 5 years of hands-on experience implementing DSP algorithms in production
systems - Proven experience implementing and optimizing DSP software on architectures such as
VLIW and SIMD/vector processors (and/or ARM-based platforms), including
performance-critical kernel development - Experience with low-level performance optimization techniques, including use of chip-
specific intrinsics and analysis of compiler-generated assembly code - Experience developing real-time embedded software in multi-core / multi-processor
environments - Experience working with MATLAB and Python for algorithm development, modeling,
validation, or prototyping
Preferred Qualifications
- Experience with specific DSP platforms and toolchains (e.g., Xtensa, Hexagon, CEVA,
or proprietary VLIW/vector DSP processors) - Experience implementing high-rate signal processing pipelines for sensor or perception
systems (radar, lidar, camera, audio, or similar) - Strong familiarity with RTOS environments, real-time scheduling, and synchronization
primitives - Understanding of SoC-level integration topics such as memory hierarchy, DMA engines,
cache behavior, and accelerator interfaces - Experience building automated testing, CI/regression frameworks, and performance
regression tracking for embedded DSP software
Benefits (subject to location and local regulations)
- Competitive salary and equity
- Comprehensive medical, dental, and vision coverage
- 401(k) retirement plan
- Flexible vacation and time-off policy
- Collaborative, fast-paced, and inclusive work environment
- Opportunity to work on cutting-edge technologies with a highly cross-functional team
