- Location
- Taipei, Taipei City,TW, TW
- Type
- Full-time
- Department
- Design
- Seniority
- Senior
- Education
- PhD
- Source
- Eightfold
Description
1. Architect DRAM access pipelines for ISP and Video workloads, including read/write scheduling, burst shaping, buffering, and scalable support for high resolution and multi stream scenarios. 2. Design and optimize arbitration and QoS mechanisms—such as priority based, aging, credit/token based, fairness, and deadline/urgency policies—to ensure deterministic latency for real time imaging and video paths. 3. Develop performance and traffic models (C/C++/SystemC/Python) to analyze DRAM bandwidth, latency, utilization, and to evaluate arbitration and QoS strategies across representative workloads. 4. Collaborate cross functionally with ISP/Video architecture, NoC/interconnect, and memory controller teams to align system level dataflows, bandwidth budgets, and runtime QoS requirements. 5. Support verification and post silicon validation, including DRAM trace analysis, bandwidth/latency profiling, QoS tuning, issue debug, and preparation of architecture/performance documentation. Master's degree in Electrical Engineering, Computer Engineering, Computer Science, or related field. Strong understanding of DRAM systems (LPDDR4/5), memory controller scheduling, row/bank/refresh behavior, and SoC memory hierarchy. Solid knowledge of ISP and Video dataflow patterns, including RAW/YUV access, multi frame fusion, motion/ME pipelines, and video reference frame behavior. Proficiency in RTL micro architecture, performance analysis, and modeling using SystemVerilog, SystemC, C++, and/or Python. Experience analyzing bandwidth, latency, utilization, and backpressure effects in complex SoC environments. Demonstrated ability to drive designs from concept through modeling, implementation, and silicon validation. Experience with DRAM access design for ISP or video pipelines such as HDR, noise reduction, rotation/warp, or video encoder/decoder flows. Background in NoC/MC arbitration, QoS tuning, bandwidth shaping, and pre silicon performance modeling. Familiarity with multi subsystem concurrency across camera, video, display, CPU/GPU/AI pipelines and system level bandwidth planning. Experience with post silicon performance profiling, QoS analysis, and DRAM trace based debugging. Strong cross team communication skills and the ability to work across ISP/Video/Display/ML architecture groups. Bachelor's degree in Science, Engineering, or related field and 2+ years of ASIC design, verification, validation, integration, or related work experience. OR Master's degree in Science, Engineering, or related field and 1+ year of ASIC design, verification, validation, integration, or related work experience. OR PhD in Science, Engineering, or related field.