Sr Staff, Physical Design Architect (Flow & ML Automation)

Job Description:

Overview

As the RTL-to-GDS Flow Lead, you will be the primary architect of the physical implementation backbone for our next-generation, high-speed CPU cores. This is a high-visibility role focused on pioneering automated, ML-driven methodologies that push Power, Performance, and Area (PPA) beyond industry standards on N3 and N2 process nodes. You will bridge the gap between architectural intent and physical reality, ensuring predictable, world-class execution.

Key Responsibilities

• Methodology Leadership: Architect, validate, and deploy “Best-of-Class” RTL-to-GDS recipes tailored for high-frequency CPU implementation.
• ML-Driven Optimization: Lead the integration and scaling of AI/ML optimization tools (e.g., Synopsys DSO.ai) to autonomously navigate the massive PPA design space.
• Strategic EDA Engagement: Serve as the primary technical interface with EDA partners (Synopsys) and IP providers to influence tool roadmaps and co-optimize Standard Cell/SRAM libraries for PPA.
• Hierarchical Strategy: Define the global floor planning, budgeting, and integration strategies for multi-million gate, high-speed hierarchical designs.
• Predictability & Scale: Design and maintain robust build systems, automated dashboards, and execution checklists to ensure “push-button” predictability and minimize manual intervention.
• PPA Modeling: Early-stage PPA estimation and pathfinding to influence RTL and architectural trade-offs.

Requirements

• Expertise: 10+ years of experience in Physical Design with a proven track record of closing high-speed CPUs (3GHz+).
• Education Qualification: A Bachelor’s, Master’s, or PhD in E&E, ECE, CS, or VLSI is required.
• Advanced Nodes: Deep, hands-on experience with TSMC N3 is required; exposure to N2 is highly preferred.
• Tool Mastery: Power-user knowledge of the Synopsys Fusion Design Platform (Fusion Compiler, IC Compiler II, PrimeTime).
• AI/ML Integration: Demonstrated experience deploying ML-based design space optimization tools in a production environment.
• Systemic Automation: Expert-level proficiency in Python, Tcl, or Perl, with a mindset for building scalable, maintainable “Flows-as-Code.”
• Analytical Depth: Deep understanding of timing closure, IR drop, electromigration (EM), and crosstalk challenges unique to high-performance cores.