Senior Design Engineer - High Speed Cache
TachyumJob Title
Senior Design Engineer - High Speed Cache
Role Summary
Design and implement a high-performance L2 cache unit serving AI processing elements. The role sits on the hardware/ASIC design team and focuses on cache controller, coherency, and low-power logic for multicore SoCs.
The position emphasizes RTL and physical-aware design practices to meet performance, power, and timing targets.
Experience Level
Senior level. The posting requests 2β4 years of relevant experience; exceptional candidates with less experience may be considered.
Responsibilities
Key responsibilities include:
- Design and implement L2 cache and cache controller logic for high-performance AI processors.
- Develop and verify cache coherency protocols and cache hierarchy behavior.
- Create RTL in Verilog/SystemVerilog and support synthesis and static timing closure.
- Perform STA, CDC analysis, linting, and timing-driven design iterations.
- Apply low-power and deep-submicron design techniques to meet power and performance targets.
- Collaborate with microarchitecture, verification, and physical design teams to integrate cache into multicore SoC.
- Optimize cache/memory interfaces for bandwidth, latency, and coherence.
Requirements
Must-have technical skills and experience:
- 2β4 years of relevant ASIC/SoC design experience (consideration for exceptional candidates with lower experience).
- Strong understanding of computer architecture and multicore processor pipelines.
- Experience with cache controller design and cache coherency protocols.
- RTL design experience (Verilog/SystemVerilog), synthesis, STA, CDC, and lint flows.
- Experience with logic design in deep-submicron technologies and low-power techniques.
Nice-to-have:
- Familiarity with ARM and x86 processor designs.
- Programming and scripting skills (C, Perl, shell, Python, AWK).
Education Requirements
Not specified.
About the Company
Company: Tachyum
Headquarters: Sunnyvale, CA, USA
Tachyum develops Prodigy, a universal processor that combines CPU, GPU, and TPU capabilities to deliver high-performance, energy-efficient computing for AI, HPC, and cloud data centers. The company focuses on ultra-low power, hyperscale deployments and operates offices in the United States and Slovakia.
