Staff Silicon Physical Design Engineer - Bengaluru

About us

Graphcore is a globally recognised leader in Artificial Intelligence computing systems. The company designs advanced semiconductors and data centre hardware that provide the specialised processing power needed to drive AI innovation, while delivering the efficiency required to support its broader adoption.

As part of the SoftBank Group, Graphcore is a member of an elite family of companies responsible for some of the world’s most transformative technologies.

Job Summary

We are looking for high-quality silicon physical design engineers to complement our existing exceptional team. We have a range of roles available with focus on those with extensive ranges of skills and experience although exceptional candidates with less experience will be considered. We want people who work collaboratively and proactively within a team focusing on collectively achieving our goals and creating the right engineering solutions. Good communication is essential, as is the ability to adapt and learn – we value the right characteristics more than specific experience.

For the successful candidate we offer an open, honest and collaborative environment working on leading-edge designs at the most advanced nodes. Our engineers are not siloed, and they are trusted and encouraged to take ownership of their designs and problem solutions. You will become part of a team that looks for improvements to everything we do: our designs, our flows, our methodologies, our infrastructure.

The Team

The physical design team sits within the wider silicon design team which includes RTL, verification and DFT and with whom we collaborate extensively. Our work additionally involves strong links with architecture, packaging and product engineering. We are responsible for working with those teams to create high-quality RTL and then to build the final chip layout (e.g. GDSII) ensuring a signoff-quality design is delivered to the Foundry (e.g. TSMC).

Responsibilities and Duties

Applicants will be expected to contribute technically to the development of Graphcore's next generation of AI superchips, focusing on achieving robust, high-performance and power-efficient designs in leading-edge process technologies while meeting ambitious development schedules. Contributions are expected to span multiple areas and involve:

• using state-of-the-art EDA tools and in-house Graphcore flows to deliver final designs that are of sign-off quality

• enhancing existing flows and in-house tools/APIs to support new features and/or methodologies

• analysis and/or debugging of complex engineering problems leading to workable solutions

• developing an understanding of emerging technical issues and applying that knowledge to optimise in-house flows and methodologies

Candidates will be expected to work closely both with other teams within Graphcore and with 3rd party support engineers/contractors, ensuring good communication between all parties, and to contribute meaningfully to the overall efficiency and success of the Physical Team.

Candidate Profile

Essential:

• A Degree in Electronic/Electrical Engineering, Computer Science or related subject

• Be highly motivated, a self-starter, and a team player

• Enjoy taking responsibility and improving skills and knowledge

• Excellent problem-solving skills for debugging issues seen and finding root causes

• Ability to program/script (required to solve design issues, typically in Tcl and Python)

• Experience in 7nm or smaller technologies

• A good breadth of experience with physical design flows including: Floorplanning/Budgeting, Synthesis, Place and Route, Clock CTS, Timing Analysis, Logical Equivalence, Physical Verification (DRC/LVS/ERC)

• Minimum 6/7 years relevant engineering experience

Desirable

• In depth expertise in one or more aspects of physical design flows

• Structuring of builds to maximise PPA

• Silicon transistor knowledge including std cell libraries and/or memories

• 2D and 3D design (CoWoS, UCIe etc.)

• High speed, high power and/or full reticle chip designs

• Experience with Ethernet, PCIe, LPDDR, HBM, UCIe implementations

• Power integrity and optimization

• 2nm or 3nm technologies

• Chip finishing (pad rings, chip level LVS/DRC/ERC)

• Design for test

• Team management

• Technical leadership

• Project Planning