Design Verification Architect

• No endless meetings, no PowerPoint presentations to convince middle management., We are seeking a highly experienced Design Verification Architect to lead the definition and execution of scalable, reusable, and high-quality verification strategies for next-generation SoCs targeting AI, HPC, and Networking applications. This role requires deep expertise in verifying large and complex SoCs and a passion for building next-gen verification platforms, leveraging automation, portability, and AI-driven methodologies.

What You’ll Do

• Define and drive the overall verification architecture for large-scale SoCs, ensuring scalability from IP * subsystem * full SoC level
• Architect and develop reusable, modular, and portable verification platforms that can be leveraged across multiple chip programs
• Establish methodologies for seamless reuse and portability, including Portable Stimulus (PSS)-based verification flows
• Lead the development of high-performance, coverage-driven testbenches using industry-standard methodologies (e.g., UVM)
• Drive multi-level verification strategies, enabling efficient scaling across:
• Block-level, Multi-block/subsystem level, Full-chip SoC level
• Define and enforce verification quality metrics, coverage closure strategies, and sign-off criteria
• Champion a “fail-fast” philosophy to detect issues early and improve overall design quality
• Drive initiatives toward first-pass silicon success, minimizing escapes and post-silicon debug effort
• Leverage AI/ML techniques and AI agents to:
• Automate testbench generation
• Improve stimulus generation and coverage closure
• Enhance regression efficiency and debug productivity
• Enable intelligent verification QA and anomaly detection
• Collaborate closely with design, architecture, physical design, and software teams to ensure alignment and early issue detection
• Mentor and guide DV teams on best practices, methodology adoption, and architectural decisions

If you’re reading this, you’re probably comfortable. You have a good job at a stable company with all the benefits., * Bachelor’s/Master’s degree in Electrical Engineering, Computer Engineering, or related field

• 12+ years of experience in Design Verification, with significant exposure to large SoC verification
• Proven experience in defining and deploying verification architectures for complex chips
• Strong expertise in:
• SystemVerilog and UVM
• Coverage-driven verification (functional + code coverage)
• Assertion-based verification (SVA)
• Hands-on experience with Portable Stimulus (PSS) and cross-level stimulus reuse
• Deep understanding of SoC architectures in AI, HPC, or Networking domains
• Experience with multi-IP integration challenges, coherency, high-speed interfaces, and large-scale data movement
• Strong debugging skills and ability to handle complex system-level issues, * Experience with AI/ML applications in verification workflows
• Exposure to emulation, prototyping, or hybrid verification platforms
• Familiarity with performance verification and power-aware verification
• Knowledge of industry-standard protocols (e.g., PCIe, DDR, Ethernet, CXL, UCIe)
• Experience building verification frameworks reused across multiple tape-outs

Key Attributes

• Strong architectural thinking with a systems-level mindset
• High ownership with a quality-first approach
• Bias toward automation, reuse, and scalability
• Passion for adopting cutting-edge technologies, especially AI-driven verification
• Ability to operate in a fast-paced environment with a fail-fast, learn-fast attitude
• Excellent collaboration and leadership skills

Success in This Role Looks Like

• High degree of verification reuse across programs
• Achievement of coverage and quality goals within schedule
• Reduction in verification cycle time through automation and AI
• Strong track record of first-time silicon success
• Minimal post-silicon escapes and debug cycles

The AI infrastructure market is exploding. Every hyperscaler, every cloud provider, every AI company is building custom silicon. But they all face the same problem: how do you connect hundreds of chips, deliver clean power at scale, and move terabits of data without melting the package?

That’s what we solve. TylSemi builds the chiplet infrastructure IP - the IO, power delivery, and interconnect building blocks - that makes AI/HPC systems actually work at scale.