Research Assistant and PhD student at Freie Universität Berlin, building efficient, deployable machine learning systems for resource-constrained devices.

Google Scholar LinkedIn CV

	Open to collaboration
Field	TinyML / MLOps Systems
Stack	C - Rust - IREE - LLVM/MLIR - TVM
Location	Berlin, Germany

Email: zhaolan.huang[at]fu-berlin.de / zhalan94[at]gmail.com

Career

Freie Universität Berlin (2023 — Present)

Research Assistant @ Berlin, Germany

• Ariel-ML: Rust-based TinyML framework featuring multi-core scheduling on microcontrollers.
• TinyDéjàVu: Eliminate up to 90% RAM usage on streaming inference.
• msf-CNN: CNN fusion technique that achieves inference using 50% less RAM on MCUs.
• TinyChirp: Bio-acoustic TinyML pipeline that extended sensor lifetime from 2 to 18 weeks.
• RIOT-ML: Secure CI/CD for ML models on low-power devices based on IETF SUIT.
• U-TOE: TinyML benchmarking toolkit with RIOT OS support.

MedRhein GmbH (2021 - 2022)

Product Manager & SAP Consultant @ Berlin, Germany

• Led and coordinated the implementation of SAP Business One.
• Designed prototypes for internal ERP tools and the company website.

Guangdong University of Technology (2014 - 2018)

Embedded Systems Engineer (freelance) @ Guangzhou, China

• Developed software for bedside multiparameter and anesthesia monitors.
• Developed a cardiopulmonary resuscitation training system.

Publications [Full list]

• Ariel-ML: Embedded Rust Leveraging Multicore for Neural Networks on Heterogeneous Microcontrollers
  Z. Huang, K. Schleiser, G. Myung, E. Baccelli
  ISIoT 2026 / arXiv / Code

• TinyDéjàVu: Smaller Memory Footprint & Faster Inference on Sensor Data Streams with Always-On Microcontrollers
  Z. Huang and E. Baccelli
  ISIoT 2026 / arXiv / Code

• msf-CNN: Patch-based Multi-Stage Fusion with Convolutional Neural Networks for TinyML
  Z. Huang and E. Baccelli
  NeurIPS 2025 / arXiv / Code

• TinyChirp: Bird Song Recognition Using TinyML Models on Low-power Wireless Acoustic Sensors
  Z. Huang, A. Tousnakhoff, P. Kozyr, R. Rehausen, F. Bießmann, R. Lachlan, C. Adjih, E. Baccelli
  IEEE IS2 2024 / arXiv / Code

• RIOT-ML: toolkit for over-the-air secure updates and performance evaluation of TinyML models
  Z. Huang, K. Zandberg, K. Schleiser, and E. Baccelli
  Annals of Telecommunications / Code

• U-TOE: Universal TinyML On-board Evaluation Toolkit for Low-Power IoT
  Z. Huang, K. Zandberg, K. Schleiser, and E. Baccelli
  PEMWN 2023 / arXiv / Code

Education

Freie Universität Berlin (2023 — Present)

PhD Student in Computer Science @ Berlin, Germany

Machine learning systems; model optimization and CI/CD on resource-constrained devices.

Supervisor: Prof. Dr. Emmanuel Baccelli

Technische Universität Darmstadt (2019 - 2021)

M.Sc. Electrical Engineering and Information Technology - Automatic Systems @ Darmstadt, Germany

Thesis: Pattern Recognition in the Capacitive Electrocardiogram and Reconstruction of the Reference Electrocardiogram.

Guangdong University of Technology (2013 - 2017)

B.Eng. Automation @ Guangzhou, China

(Natural) Languages

Language	Level
Chinese	Native
German	C1 - Professional
English	Professional