I’m a professor in electronic system design at KTH Royal Institute of Technology.

My research interests are in energy-efficient custom computing machines and their design automation. The custom computing machines mainly target CMOS based logic. Besides CMOS, I am also exploring computation in memristors to eliminate the Von-Neumann bottleneck. Finally, some preliminary work is planned for wave-based computing.

A novel Synchoros VLSI Design style has been proposed to automate the design of custom computing machines. The word synchoros is derived from the Greek word choros for space. Synchoricity is the spatial analogue of the temporal concept of synchronicity. Based on synchoros VLSI Design style, a design framework inspired by the Lego toys called SiLago (Silicon Lego) is being developed. See SiLago for more details.

The main advantages of SiLago are:

  • Enables achieving ASIC comparable computational efficiency,
  • Non-incrementally lowers the design cost of complex applications and systems. The design automation flow generates DRC and timing clean GDSII macro without the end-user having to use logic or physical syntheses
  • It has the potential to lower the manufacturing cost

The SiLago framework is being applied to two industrial edge AI applications: neuromorphic computing and genomics. The SiLago framework is also planned to be enhanced for designing computation in memristors and wave-based computing.