Somdip Dey

Fixed Term Teacher
School of Computer Science and Electronic Engineering (CSEE)
Assistant Lecturer
School of Computer Science and Electronic Engineering (CSEE)
Postgraduate Research Student
School of Computer Science and Electronic Engineering
 Somdip Dey


Ask me about
  • Embedded Machine Learning
  • Embedded AI
  • On-Device AI
  • Computer Vision
  • Run-time Management of Computing Resources
  • Object Oriented Programming Languages (C#, Java)
  • Information Security and Privacy


Somdip received his first class B.Sc. (Honours) degree in Computer Science from St. Xavier's College (Autonomous), Kolkata, under the University of Calcutta. From his second year during his undergraduate studies he started pursuing research in Computer Security, which gave him the opportunity to publish and present his work at several conferences and journals. During his undergraduate studies he has also worked at Microsoft as a Student Partner, developing applications for Windows Phone platform. After completing his bachelor's he came to the UK to pursue his master's (M.Sc.) in Advanced Computer Science with specialisation in Computer Systems Engineering at the University of Manchester. His master's thesis was on "Efficient Data Input/Output (I/O) For Finite Difference Time Domain (FDTD) Computation On Graphics Processing Unit (GPU)". Upon graduating he joined Steanne Solutions Ltd. as a Software Development Engineer, where he led the development of the flagship product of the company. He has also co-founded Codeepy U.K. Pvt. Ltd., which helps internationals in the U.K. to find sponsored work, and ReMe Basket Ltd., which helps an individual to reduce food waste. Somdip's current research is focused on "Energy Efficient and Reliable Computer Vision Processing on Multi-core Processors" under the supervision of Professor Klaus McDonald-Maier and Dr Amit Singh.


  • B.Sc. (Honours) in Computer Science University of Calcutta (2012)

  • M.Sc. in Advanced Computer Science with specialization in Computer Systems Engineering The University of Manchester (2014)

Research and professional activities


New Approaches To DVFS In Mobile MPSoC

This project aims to address energy and reliability concerns in modern multi-core processors while running computer vision applications on them. This needs significant research efforts in power-aware and temperature-aware run-time management (RTM) of computing resources, such as CPUs, GPUs and RAMs present in modern multi-core processors. The role of the project involves becoming familiar with the state-of-the-art research relevant to the topic and then devising algorithms to find appropriate

Supervisor: Klaus McDonald-Maier , Amit Singh



Colchester Campus

Working pattern:

Work now and I can sleep after I am dead