Junior/PhD Researcher in Brain-Inspired Edge AI Processors

Design the next generation of low-power AI accelerators for the edge 

Overview

Imec Netherlands has established a world-leading track record on the design of novel sensors for human health applications and ultra-low-power IoT devices in general. For instance, our teams continue to develop innovations in implantable therapeutics, medical/wearable devices, environmental monitoring, accurate localization, and many other fields. At our HQ in Leuven, Belgium we are world-renowned for pioneering new process technologies in collaboration with industrial foundries and with the broader semiconductor ecosystem. imec’s research tracks enable unique opportunities for end-to-end innovations. 

The team at imec Netherlands develops innovative solutions for both the Life Science and Medical Device domain, as well as large-scale, heterogeneous sensor networks that connect billions of devices within the Internet of Things (IoT). We have built various ultra-low-power (ULP) radio ICs, including Wi-Fi, Bluetooth-LE, and a UWB 6-10 GHz radio tuned towards precise ranging and localization. Additionally, we are advancing a wide range of neuro-inspired architectures, efficiently implementing them in ICs across multiple computational domains, such as spiking neural networks, online and on-chip learning, sparse coding, and dynamic control. Our novel technologies will enable a wide range of smart IoT applications. Furthermore, we deliver our designs to customers, secure patents for the different inventions, and share our findings through publications in renowned conferences and journals, including ISSCC, ICC, WCNC, JSSC, and ISCAS.   

As a Junior Researcher, you will contribute to the design of digital in-memory brain-inspired edge processors for Edge AI, investigating methods, techniques, and architectural enhancements that improve energy efficiency, latency, scalability, and robustness. You will bridge AI algorithms, system-level architecture, and hardware implementation, contributing to inference-centric designs.  

What you will do

You will design and deploy Edge‑AI architectures for constrained platforms, developing and optimizing emerging models (e.g., transformers, sparse and quantized networks) on edge hardware. You will drive dataflow, memory, and HW/SW co‑design in close collaboration with IC, sensor, and system teams, build end‑to‑end demonstrators, and contribute to research. Your responsibilities as a Junior/PhD Researcher include:   

Key Responsibilities:

• Investigate architectural enhancements for digital in-memory and near-memory AI processing, with emphasis on energy efficiency, latency, scalability, and robustness. 
• Contribute to the system-level architecture exploration of next-generation brain-inspired Edge AI processors, including digital in-memory and near-memory accelerator concepts. 
• Investigate and design novel digital macros, accelerator building blocks, or memory-centric compute units for efficient neural network processing. 
• Explore quantization, model compression, and algorithm-hardware co-optimization strategies for efficient neural network deployment. 
• Work on applications such as multimodal sensor fusion on digital accelerators based on advanced nanoelectronics technologies. 
• Collaborate with imec researchers, TU Delft supervisors, and international project partners within the MSCA TIRAMISU network. 
• Carry out planned secondments with European project partners as required by the project. 
• Contribute to scientific publications, project deliverables, technical documentation, and dissemination activities. 
• Support organizational and coordination tasks relevant to the project and imec’s research activities. 

Required Qualifications:

• Applicants must comply with the MSCA Doctoral Network mobility rule: they must not have resided or carried out their main activity, such as work or studies, in the Netherlands for more than 12 months in the 36 months immediately before the recruitment date. 
• Master’s degree, or equivalent, in Computer Engineering, Electrical Engineering, Computer Science, or a closely related field. 
• Clear interest in digital in-memory and near-memory computing, brain-inspired computing, Edge AI, embedded systems, or machine learning. 
• Basic understanding of system design, computer architecture, and machine learning concepts. 
• Good skills in VHDL or Verilog. 
• Ability to code in C/C++ or Python. 
• English language proficiency and strong writing and communication skills suitable for an entry-level research position. 
• Capacity to work both independently and as part of an international, multidisciplinary research team. 

Preferred Experience:

• Research or professional experience in embedded systems, digital hardware design, machine learning, or hardware/software co-design. 
• Familiarity with RISC-V architecture, FPGA development, or EDA tools. 
• Familiarity with ML algorithms, DNN architectures, spiking neural networks, quantized neural networks, or multimodal sensor fusion. 
• Previous research publications, open-source contributions, or demonstrable projects in relevant domains. 

Project context:
The position is part of the MSCA TIRAMISU project. The candidate will be employed by imec for a defined period of approximately two years (exactly 23.5 months) and will be enrolled as a PhD student at TU Delft. The contract is expected to end after this period. If imec or its partners secure additional funding, an extension may be considered based on the candidate’s performance and project needs. The candidate is expected to perform secondments with European project partners as part of the doctoral network. 

What we do for you

As a PhD Researcher at imec the Netherlands you get the chance to scale future computing beyond what is possible with silicon CMOS. In collaboration with your colleagues, you will work on a number of challenges to the design of efficient (brain-like) chips: scalable, energy-efficient, highly flexible architectures that can learn with incredible efficiency.   

Imec The Netherlands, as an employer of choice, offers you an attractive compensation and benefits package including a 100% employer-paid pension plan and flexible working hours. Furthermore, imec-NL offers a wide range of training possibilities. For international candidates, we provide support in the relocation process.   

Interested

Does this position sound like an interesting next step in your career at imec? Don’t hesitate to submit your application by clicking on ‘APPLY NOW’.
Should you have more questions about the job, you can contact jobs@imec.nl.