tinyML: Deep Learning Models on Low-power Micro-Controllers

Discover how TinyML brings deep learning directly onto microcontrollers and sensors. This tutorial guides embedded developers through the full pipeline, from data prep and lightweight model design to evaluation and code generation with TVM, enabling intelligent, low-power edge applications.

Intelligence for the IoT requires to stream huge amounts of data from edge sensors towards the cloud, where deep learning models interpret the data. EdgeAI moves deep learning models from the cloud onto the Edge platforms themselves offering huge gains in terms of connectivity requirements, energy, cost, privacy and end-to-end latency. tinyML or Extreme Edge AI moves the deep learning tasks even further right onto the microcontrollers connected to the sensors.

This class provides an in-depth introduction to TinyML, the field of deploying machine learning (ML) models on resource-constrained edge devices like microcontrollers and sensors. Participants will learn the full pipeline of developing efficient ML models optimized for low-power, low-memory environments. The class covers the essential stages: model preparation, where participants will explore data preprocessing, feature selection, and designing lightweight models suitable for edge hardware; model evaluation, focusing on performance metrics, accuracy trade-offs, and model compression techniques to ensure feasibility on small devices as well as evaluation in the context of embedded applications; and finally, code generation, where participants will learn how to convert models into deployable code using tools like TensorFlow Lite or TVM, generating firmware for real-world applications.  
 

Agenda & Content

For a detailed timetable and additional information, please see Agenda & Content in the left menu.

Target audience, eligibility & prices

Embedded Software Developers that want to learn about how to integrate deep learning into embedded applications

This course is open and free of charge for all participants from academia, industry, and public administration from EU and/or EuroHPC JU member countries.

Entry level & prerequisites

Intermediate – Some python experience and experience in embedded software development is required. 

Course format

This course will be delivered as a HYBRID Course via ZOOM ONLINE / LIVE @ AI Factory Austria AI:AT, Karl-Farkas-Gasse 22, 1030 Wien 

Lecturer

Daniel Mueller-Gritschneder (TU Wien)                      
Daniel Mueller-Gritschneder is a full professor at the Institute of Computer Engineering at TU Wien Informatics, Austria, a position he has held since 2024. Prior to this, he served as a research group leader and acting professor for Real-Time Systems at the Technical University of Munich (TUM), Germany. He received his Dipl.-Ing., Dr.-Ing., and habilitation degrees from TUM in 2003, 2009, and 2019, respectively. Throughout his career, he has contributed to numerous collaborative research projects in TinyML and Edge AI, working closely with industry partners including Infineon, Bosch, and BMW. His research focuses on the optimization of deep learning models and deployment toolchains for embedded platforms, spanning tiny microcontrollers to high-performance central compute units. He is a regular program committee member for leading EDA conferences such as DAC, ICCAD, DATE, SAMOS, and CODES+ISSS, and is active in the RISC-V community, including contributions to the RISC-V Summit Europe. Recent works he co-authored were recognized with the best paper award at the SAIAD Workshop of CVPR2025, the CODAI workshop 2023, as well as the Best Student Paper Award at the WACV 2024. 

Philipp van Kempen (TU München)   
Philipp van Kempen works in the Electronic System Level (ESL) group at the TUM Chair for Electronic Design Automation (EDA), developing ML and software compiler solutions to tackle deployment challenges for Extreme Edge AI (TinyML) applications. His research focuses on automatic retargeting of embedded workloads to customized RISC-V platforms. He also has a strong background in virtual prototyping and a passion for open-source ecosystems. 

Language

English

Date, time, and location

13.05.2026, 09:30 – 12:30 (CET), via ZOOM ONLINE / LIVE @ AI Factory Austria AI:AT, Karl-Farkas-Gasse 22, 1030 Wien 

Registration

Registration is required, the registration form can be found on top of this page in the menu on the left.

Please register with your official institutional email address to prove your affiliation.

You will get an automatic confirmation by email (subject starting with "[Indico] Registration"), please check your Spam/Junk folders.

Following your successful registration, you will receive further information a few days before the course.

Please do not hesitate to contact us at training@ai-at.eu if you have any questions.

Waitinglist

After the number of registrations has reached its maximum or the registration form has been closed, you may want to send us an email (training@ai-at.eu) stating that you are interested to be put on the waiting list (vacancies may occur due to cancelations, etc.).

Modification, withdrawal & no-show policy

Your registration is binding. Please only register for the course if you are really going to attend.

You can update your registration data or withdraw your registration anytime before the registration form has been closed via the link "Manage my registration" which you can find at the bottom of your automatic email confirmation (subject starting with "[Indico] Registration").

Alternatively, or after the registration form has been closed, please inform us about your cancelation or any change in your registration data (especially your mobile-phone number) via email (training@ai-at.eu).

No-show policy: If you do not cancel and do not show up at the course you will be blacklisted and excluded from future training events.

Organizers

This course is organized by AI Factory Austria AI:AT.

Acknowledgements

AI Factory Austria AI:AT has received funding from the European High-Performance Computing Joint Undertaking (JU) under grant agreement No 101253078. The JU receives support from the Horizon Europe Programm of the European Union and Austria (BMIMI / FFG).