CV
Zain Taufique

Zain Taufique

PhD Researcher · University of Turku
github scholar dblp orcid LinkedIN
whoami

I am a PhD researcher at the University of Turku, working on efficient execution of AI workloads on heterogeneous computing platforms. My research focuses on runtime systems and optimization techniques for multi-DNN pipelines and transformer inference on edge devices.

My work studies how modern AI models can be scheduled and optimized across CPU, GPU and NPU systems while minimizing latency, improving throughput and reducing energy consumption.

My earlier work spans neural signal processing, low-power IC design, and approximate computing for wearable biomedical devices.

ping zain
email
news
Nov 2025 Presented Twill paper at AI Day by FCAI, Helsinki, Finland
Oct 2025 Presented Twill paper at the IEEE ICCAD conference, Munich, Germany
Mar 2025 Presented HiDP paper at the IEEE DATE conference, Lyon, France
Nov 2024 Presented TANGO at the IEEE ICCD conference, Milan, Italy.
Jan 2024 Presented distributed inference paper at the IEEE ASP-DAC conference, Incheon, South Korea.
cat publications.bib
2025
DATE
HiDP: Hierarchical DNN Partitioning for Distributed Inference on Heterogeneous Edge Platforms
Z Taufique, A Vyas, A Miele, P Liljeberg, A Kanduri
Design, Automation and Test in Europe (DATE)
DOI
ICCAD
Twill: Scheduling Compound AI Systems on Heterogeneous Mobile Edge Platforms
Z Taufique, A Vyas, A Miele, P Liljeberg, A Kanduri
IEEE/ACM International Conference on Computer-Aided Design (ICCAD)
Paper
ACM TECS
Exploiting Approximation for Run-time Resource Management of Embedded HMPs
Z Taufique, A Kanduri, A Miele, A Rehmani, C Bolchini, N Dutt, P Liljeberg
ACM Transactions on Embedded Computing Systems
DOI
2024
J. Sys. Arch.
Adaptive Approximate Computing in Edge AI and IoT Applications: A Review
HJ Damsgaard, A Grenier, D Katare, Z Taufique, S Shakibhamedan
Journal of Systems Architecture
DOI
ICCD
TANGO: Low Latency Multi-DNN Inference on Heterogeneous Edge Platforms
Z Taufique, A Vyas, A Miele, P Liljeberg, A Kanduri
IEEE International Conference on Computer Design (ICCD)
DOI
ASP-DAC
Adaptive Workload Distribution for Accuracy-Aware DNN Inference on Collaborative Edge Platforms
Z Taufique, A Miele, P Liljeberg, A Kanduri
Asia and South Pacific Design Automation Conference (ASP-DAC)
DOI
arXiv
Characterizing Accuracy Trade-offs of EEG Applications on Embedded HMPs
Z Taufique, MAB Altaf, A Miele, P Liljeberg, A Kanduri
arXiv preprint
Paper
2021
IEEE TCAS-II
A Low Power Multi-Class Migraine Detection Processor Based on Somatosensory Evoked Potentials
Z Taufique, B Zhu, G Coppola, M Shoaran, MAB Altaf
IEEE Transactions on Circuits and Systems II
Paper
NORCAS
Approximate Feature Extraction for Low Power Epileptic Seizure Prediction in Wearable Devices
Z Taufique, A Kanduri, MAB Altaf, P Liljeberg
IEEE NORCAS
Paper
A-SSCC
An 8.7 µJ/class FFT Accelerator and DNN-Based Configurable SoC for Multi-Class Chronic Neurological Disorder Detection
Z Taufique, B Zhu, G Coppola, M Shoaran, W Saadeh, MAB Altaf
IEEE Asian Solid-State Circuits Conference (A-SSCC)
Paper
2019
BioCAS
An ECG Processor for the Detection of Eight Cardiac Arrhythmias with Minimum False Alarms
MA Sohail, Z Taufique, SM Abubakar, W Saadeh, MAB Altaf
IEEE BioCAS
Paper