Aktiviere Job-Benachrichtigungen per E-Mail!

PhD in digital accelerator design for deep learning and neuromorphic computing

Fraunhofer-Gesellschaft

Dresden

Vor Ort

EUR 40.000 - 55.000

Vollzeit

Vor 5 Tagen
Sei unter den ersten Bewerbenden

Erhöhe deine Chancen auf ein Interview

Erstelle einen auf die Position zugeschnittenen Lebenslauf, um deine Erfolgsquote zu erhöhen.

Zusammenfassung

A leading research institute is seeking a doctoral student to develop innovative AI hardware platforms. The role involves RTL design, FPGA integration, and creating research publications. Candidates should have a master's degree in electrical engineering or a related field, along with experience in FPGA design. The position offers professional development opportunities and a supportive work environment.

Leistungen

Flexible working hours
Remote options
Career development initiatives

Qualifikationen

  • Completed university studies in electrical engineering or comparable.
  • Experience in FPGA design for different technologies and suppliers.

Aufgaben

  • Responsible for RTL design of digital blocks and system-level integration.
  • Create and execute test benches for RTL and timing simulations.

Kenntnisse

Problem-solving
Analytical skills
Adaptability
English

Ausbildung

Master's degree in Electrical Engineering
Diploma in Communication Engineering

Tools

VHDL
Verilog
Design Compiler
Cadence Innovus

Jobbeschreibung

Everyone is talking about artificial intelligence. But who is developing the necessary chips? We are, for example!

Would you like to help drive the development of a new highly efficient AI hardware platform as a doctoral student? At Fraunhofer IPMS, in collaboration with renowned German and European partners from science and industry, we are developing analog accelerators using novel non-volatile memory chips! By merging logic and memory elements, costs and wafer area are minimized, while energy efficiency and speed are maximized. Specifically, in this project, we are using ferroelectric memories, which can calculate AI algorithms from the field of deep learning in resistive crossbar structures with extremely low power consumption and high speed.

Furthermore, we investigate and develop innovative memory solutions in advanced CMOS technologies such as FDSOI. The development of integrated circuits also plays an important role in making these networked devices and their sensors locally intelligent. Alongside our activities in sensor platform development, the institute is strengthening its expertise in embedded machine learning, neuromorphic hardware, and deep learning accelerators.

Want to get more information? Click here.

What you will do
  • Responsible for RTL design (VHDL, Verilog) of digital blocks and their system-level integration
  • Develop design architecture and break down requirements into functional blocks
  • Create and execute test benches for RTL and timing simulations
  • Perform formal verification of FPGA designs, including timing analysis, code coverage, and coding rule checks
  • Support FPGA integration on target hardware
  • Create design documentation in compliance with internal and external standards
  • Write research papers for journals and conferences
What you bring to the table
  • Completed university studies in electrical engineering, electronics, communication engineering, information technology, or comparable, with a master's or diploma degree
  • Experience in FPGA design for different technologies and suppliers like Intel PSG (Altera), Xilinx, or Lattice
  • Knowledge of synthesis and physical design tools such as Design Compiler and Cadence Innovus is an advantage
  • Expertise in at least one of the following areas is a plus:
    • Deep learning
    • Hardware development
    • Edge AI
    • Memory technology
  • Strong problem-solving and analytical skills with the ability to lead complex design efforts
  • Quick adaptability to new technical and scientific contexts
  • Motivated with a focus on producing high-quality scientific publications and presentations at international conferences
  • Good written and spoken English skills for working in an international team
What you can expect

We offer an exciting doctoral thesis with experienced supervision, focusing on your personal and professional development. Through our doctoral college, you can network with peers and benefit from seminars and lectures, advancing your career in applied research and potentially starting as a postdoc at our institute.

We also offer female scientists the opportunity to participate in our TALENTA program, a comprehensive career and development initiative with tailored qualifications. More information is available on our website.

We value diversity and promote a work environment based on appreciation and trust. We support work-life balance with flexible working hours and remote options. We emphasize thorough induction, an open environment, and opportunities for sports, music, and team events to foster team spirit.

The weekly working time is 39 hours, with part-time options available. The position is initially limited to 3 years. Employment as a postdoc is preferred.

We welcome applications from all backgrounds, including age, gender, nationality, ethnicity, religion, disability, sexual orientation, and identity. Severely disabled persons are given preference if equally qualified. Remuneration is based on the public-sector collective agreement (TVöD), with potential performance-based bonuses.

Fraunhofer plays a key role in developing future technologies and enabling their industrial application, helping shape society through scientific excellence and innovation.

Interested? Apply online now. We look forward to meeting you!

Contact

Ms. Isabell Zwinscher
Human Resources

Telephone: +49 (0)351 8823 1227

Mr. Dr. Thomas Kämpfe
Specialty Department

Telephone: +49 (0)351 2607 3215

Fraunhofer Institute for Photonic Microsystems IPMS

www.ipms.fraunhofer.de

Requisition Number: 79639

Hol dir deinen kostenlosen, vertraulichen Lebenslauf-Check.
eine PDF-, DOC-, DOCX-, ODT- oder PAGES-Datei bis zu 5 MB per Drag & Drop ablegen.