Datapath Engineer

Job Description
This is a unique opportunity to work on cutting edge math intensive logical blocks within the best in class GPU design at Intel Graphics' group. The Numerics Group (TNG) is a new applied research group focusing on all aspects of mathematical hardware design from architecting, RTL implementation & optimization, structural implementation and formal verification & validation throughout. The group works as a parallel consultancy to the ALU designers working in GPU cluster groups within 3D & Media IPs. We supplement the expertise required by analyzing architecture and APIs, all the way to creating & verifying final structural implementations with process and library involvement - providing full control over the entire development cycle of math blocks on their path to the final product.

TNG is an applied research group comprised of computer arithmetic experts; founded on the premise that deep mathematical and logical reasoning can deliver significant hardware benefits in terms of reliability, speed, area and power efficiency. As an internal consultancy we work with and enable the existing design and validation teams to achieve their goals. We work to extract the necessary requirements, phrase the right design problem, explore the design space and fully formally prove the correctness of experimental designs. Quality means that challenging legacy decisions comes with the territory, innovation & insight are prized and every bit matters.

We seek to grow our body of knowledge in all aspects of numerical hardware design. From fundamental improvement to the implementation of multiplication and addition, through to the implementation & formal verification of transcendental functions to floating-point error analysis, TNG offers and faces a myriad of fascinating opportunities & challenges. What we discover and learn we endeavor to fully utilize, automate and disseminate. We invest in internal tools and work with vendors to continuously explore ways to optimize our design processes. Our aim is for our work to contribute and participate in best in class numerical hardware design within graphics, Intel's arithmetic community and on the international computer arithmetic stage.

The Opportunity
As part of seeding the group's expansion we are looking to hire talented individuals who will share our passion for computer arithmetic and our value of quality, innovation and insight. With breadth, depth, applied research, internal customer influence & engagement and offering a unique view of Intel's GPU IP, this role offers a rare opportunity to become a true subject matter expert.

You will help us with the following responsibilities:

  • Work within a small team of world class design and verification experts.
  • Understand, crystallize and extract the requirements for mathematical hardware blocks with various architects and engineers within the 3D & Media IP teams at Intel.
  • Own, develop, diagnose & drive Datapath improvements within 3D & Media.
  • Research a variety of hardware architectures for implementing math intensive blocks.
  • Perform full formal verification on the resultant design using industry leading tools such as Cadence's Jasper and Synopsys' Hector.
  • Work with the structural group to ensure ideal library cell selection for ideal layout.
  • Investigate final netlist for both area and power to get to the best in class final product.
  • Research new techniques for implementing a vast range of mathematical operations in silicon, becoming a Datapath Expert in the process.
  • Produce patentable work in a research and invention focused environment.
  • Educate other members of the GPU group on Datapath design, mentor & upskill members of the TNG.

We grapple with the most challenging questions in the area of mathematical hardware design. Intel Graphics Datapath differentiation rests on innovation at all levels of composition and implementation of the fundamental mathematical operations. This role requires and the group is measured on fundamental innovation.

Behavioral traits that we are looking for:
Our success hinges on understanding the needs of others, creating ideas with others, being & wanting to be understood by others, growing others as we grow ourselves. We work within the evolving diverse team that is Visual Technologies Team. Successful ideas must be generated as well as disseminated - both require equal energy.
You are:
  • Highly inquisitive
  • Excellent creative analytic problem-solving skills
  • Friendly and patient with high levels of technical empathy
  • Generous & motivational with ideas & knowledge
  • Expertly able and interested in providing complete, clear and concise delivery of complex material, whether emailed, presented, documented or verbalized tailored to the audience.
  • Consistent ability to challenge the status quo - tenacious in driving improvements & innovations
  • Ability to successfully & patiently liaise with and influence multiple groups
  • Ambitious in the quality, scope and longevity of your work
  • Proven ability and interest in mentoring others

Location & Working Environment
TNG will thrive through its countless lively whiteboard brainstorming, as such all its members will be collocated. Its primary site is Folsom.

Qualifications
We value the relentless drive to push hardware quality, the ability to create innovative Datapath architectures as well as the strong mathematical ability to verify these complex modules. The group draws strongly upon electrical engineering, computer science and mathematics. We value defining and answering the fundamental questions - even the most basic questions are a potential source of invention. To thrive, the group requires members of a range of industrial and educational experience.

Minimum Qualifications & Experience that we are looking for:
  • Masters or PhD in Mathematics, Computer Science, Electronic Engineering or equivalent and/or mixture thereof.
  • Completed academic thesis option resulting in at least one peer reviewed publication
  • At least 6 months of professional paid industrial experience
  • Experienced candidates must have worked at least 1 years in a highly numerate discipline; for example, Datapath logic design, quantitative finance, mathematical software, data mining scientist, statistician, and/or mathematical modeler.

Desirable and relevant experience would be:
  • Industrial experience in an applied mathematics role
  • Good understanding of Datapath synthesis & verification using Synopsys/Cadence tool chain
  • Experience in front-end numerical hardware RTL design and optimization
  • Understanding of floating point arithmetic and implementation
  • Formal Verification Experience
  • Verilog, SystemVerilog, C , Python, Perl, TCL experience

Are you passionate about this opportunity and ready to help us create the next generation of technologies that will transform the future for decades to come? Then we encourage you to apply and be part of our team.

Inside this Business Group
The Core and Visual Computing Group (CVCG) is responsible for the architecture, design and development of the CPU core and visual technology IPs that are central to Intel's system-on-a-chip (SoC) products and key to our datacenter, client and Internet-of-Things (IOT) platforms. CVCG strives to lead the industry through continuous innovation and world class engineering.

Posting Statement
All qualified applicants will receive consideration for employment without regard to race, color, religion, religious creed, sex, national origin, ancestry, age, physical or mental disability, medical condition, genetic information, military and veteran status, marital status, pregnancy, gender, gender expression, gender identity, sexual orientation, or any other characteristic protected by local law, regulation, or ordinance....
US JR0098442 Folsom


Meet Some of Intel's Employees

Martin S.

SoC Design Engineer

Martin uses his communication skills and technical knowledge to perform deep-level debugs of the intellectual properties that come to Intel.

Valerie P.

Principal Engineer, Data Center Group

Valerie’s job is to create a personalized network at the intersection of several technologies, from security to artificial intelligence to gaming.


Back to top