Asic Design Engineer (video Silicon Ip) - Multimedia Lab

ByteDance ByteDance · Big Tech · San Jose, CA · R&D

ASIC Design Engineer for Video Silicon IP team focusing on building efficient and scalable video codec hardware solutions (FPGA and ASIC). Responsibilities include RTL design, micro-architecture development, and collaboration with algorithm and verification teams to meet PPA targets for multi-standard codec cores.

What you'd actually do

  1. As an ASIC Design Engineer in this Video Silicon IP team, we work closely with architecture, algorithm and verification teams to build high performance and low power video processing IPs.
  2. Apply your knowledge of computer architecture and ASIC design to create ASIC design for compressing, processing still images and videos.
  3. Develop micro-architectures to meet stringent area, power, and performance (PPA) targets for multi-standard codec cores.
  4. Collaborate with algorithm teams to translate codec standard specifications and proprietary codec improvements into implementable hardware architectures.
  5. Design and implement RTL (SystemVerilog/Verilog) for video codec pipeline stages including intra/inter prediction, transform & quantization, entropy coding (CABAC/ANS), in-loop filters, and etc.

Skills

Required

  • M.S./Ph.D. in Electrical Engineering, Computer Engineering, or a related field.
  • 2 years of ASIC front-end design experience as a primary RTL owner.
  • Proficiency in SystemVerilog RTL design or High level Synthesis (HLS).
  • Familiarity with UVM/DPI/C++.
  • Solid understanding of VLSI design concepts: pipelining, clock gating, memory architecture, bus interfaces (AXI/APB), and SV assertion basics.
  • Scripting proficiency in Python for EDA flow automation.

Nice to have

  • Experience with Video Codec (H.265/HEVC, H.266/VVC, AV1, VP9, or H.264/AVC) is a big plus.
  • Experience with relevant ISP or machine learning based image/video compression.
  • Experience with FPGA prototyping for pre-silicon codec validation.
  • Practical experience using LLMs (e.g., GitHub Copilot, Claude, ChatGPT) to assist RTL design workflows.