Date posted 02/09/2026
Synopsys is the leader in engineering solutions from silicon to systems, enabling customers to rapidly innovate AI-powered products. We deliver industry-leading silicon design, IP, simulation and analysis solutions, and design services. We partner closely with our customers across a wide range of industries to maximize their R&D capability and productivity, powering innovation today that ignites the ingenuity of tomorrow.
You have spent over a decade in physical design, and somewhere along the way you realized that the next frontier is not just better tools but smarter ones. You know what it looks like when a design is stuck at 92% timing closure for three weeks, and you have the instinct to look at the data patterns, not just the methodology. Machine learning is not a buzzword to you. It is a tool you have started to explore, experiment with, and to real problems like congestion prediction, IR drop mitigation, or routing optimization.
You are equally comfortable reviewing a floorplan in Fusion Compiler and prototyping a Python script that ingests gigabytes of timing reports to surface the one thing everyone missed. You have worked on advanced nodes, 5nm, 3nm, maybe beyond, and you understand the physics that make these designs hard. You also understand that human intuition alone will not scale to solve what is coming next.
You do not just want to use AI tools. You want to build them. You want to define what intelligent automation looks like in semiconductor design, not in a lab, but in production flows that customers depend on. You ask hard questions, you push back when a model does not generalize, and you care about whether the solution actually works under real design constraints.
-
Develop AI and ML-based solutions that integrate directly with Synopsys EDA tools like Fusion Compiler, ICC2, and PrimeTime to solve PPA optimization problems across synthesis, place and route, and signoff
-
Design and deploy intelligent agent-based workflows that automate decision-making in physical design, including congestion management, timing closure, IR drop mitigation, and power density optimization
-
Build machine learning models and data pipelines using Python and big data frameworks to analyze large-scale design datasets and extract actionable insights for design convergence
-
Collaborate with customers and internal engineering teams to understand real-world design challenges at 5nm, 3nm, and future nodes, then translate those into scalable, production-ready AI-driven solutions
-
Define and document best practices for AI-driven EDA methodologies, including agent architectures, skill definitions, and orchestration frameworks that can be reused across design teams
-
Prototype and validate reinforcement learning or optimization algorithms applied to specific EDA problems like floorplanning, placement quality, or clock tree synthesis
-
Work across the full physical design flow, from RTL to signoff, ensuring AI solutions are grounded in the realities of variability, EM, IR, DRC, and DFM constraints
-
Enable design teams to close timing, power, and area targets faster by automating decisions that currently require days of manual iteration and expert judgment
-
Reduce time to tapeout for advanced node designs by building intelligent workflows that predict and resolve congestion, IR issues, and variability challenges before they become blockers
-
Define the next generation of Synopsys AI-powered design solutions that will be deployed across customer engagements globally, directly influencing how chips are designed at scale
-
Accelerate adoption of machine learning in semiconductor design by creating frameworks and methodologies that make AI accessible to physical design engineers, not just data scientists
-
Improve design quality and yield by surfacing patterns in large datasets that human analysis would miss, translating data into design decisions that matter
-
Shape the roadmap for AI integration in Synopsys EDA tools by working closely with product and R&D teams to identify high-impact use cases and validate technical feasibility
-
Help customers at the leading edge of semiconductor technology solve problems that have no established playbook, becoming a trusted technical partner in their most critical programs
-
12+ years of hands-on experience in physical design and place and route, with deep expertise across synthesis, floorplanning, placement, CTS, routing, and signoff flows
-
Strong working knowledge of Synopsys EDA tools, specifically Fusion Compiler, ICC2, or PrimeTime, and the ability to script and automate within those environments
-
Proven track record of PPA optimization on advanced nodes including 5nm, 3nm, or beyond, with direct experience addressing variability, congestion, IR drop, electromigration, and DFM challenges
Solid understanding of machine learning and artificial intelligence concepts, including supervised learning, optimization algorithms, and data-driven decision systems, with the ability to- them to EDA problems
-
Proficiency in Python and scripting for automation, data processing, and integration with EDA tools and workflows
-
Experience building or integrating AI agents, autonomous workflows, or decision systems in a technical domain, whether in EDA or adjacent fields
-
Bachelor's or Master's degree in Electrical Engineering, Computer Science, or equivalent practical experience, exposure to reinforcement learning, LLM-based tooling, agent-based architectures, or distributed data processing frameworks is a strong plus
. 12+ years of relevant experience with a degree in electrical or computer engineering or computer science.
-
You can walk into a room with a customer's design team, understand their timing closure problem in the first 20 minutes, and sketch out an AI-driven approach that feels grounded, not theoretical
-
You are comfortable with ambiguity and incomplete data, you do not wait for the perfect dataset or the perfect model, you start with what you have, iterate, and improve as you learn
-
You push back when an AI solution is being oversold or when a workflow does not account for the realities of signoff constraints, variability, or customer design rules
-
You think in systems, not just models, you understand that a great ML model is useless if it cannot be integrated into a production flow or if it breaks under real design complexity
-
You are a natural teacher and collaborator, able to explain a reinforcement learning concept to a physical design engineer and a floorplanning tradeoff to a data scientist without losing either of them
-
You care about whether your work actually ships and gets used, not just whether it works in a demo, and you are willing to do the hard work of making something production-ready
Your recruiter will
