The University of Malawi in Zomba became a vibrant centre of innovation and collaboration as it hosted a transformative Tiny Machine Learning (TinyML) workshop from 31st March to 4th April 2025, held at the Economics Building. The event was organised by the Abdus Salam

International Centre for Theoretical Physics (ICTP) and brought together researchers,

developers, students, and innovators from both local and international communities.

TinyML refers to the deployment of machine learning models directly on edge devices, such as

smartwatches, without relying on cloud computing. Two staff members—Ms Florence

Mwalwanda and Mr Stephano Mabziness Patrick—from the Centre for Resilience in Agri-Food Systems (CRAFS) participated in the workshop. They described the workshop as crucial for CRAFS, as it provided a strategic opportunity to explore how TinyML could transform

agriculture and food systems in Malawi and beyond. By supporting and actively participating in the workshop—which CRAFS also sponsored, the Centre reaffirmed its commitment to

harnessing cutting-edge technologies like edge AI to drive meaningful change in farming

communities.

Through a dynamic mix of in-depth lectures, practical lab sessions, and collaborative group

projects, participants engaged with the rapidly growing field of TinyML. This area of innovation enables artificial intelligence to operate on low-power, battery-operated devices. During the workshop, the CRAFS team learned how TinyML integrates hardware, algorithms, and software to analyse sensor data in real time at the edge, eliminating the need for cloud-based processing.

Ms Mwalwanda and Mr Patrick explored how to deploy machine learning models on

microcontrollers and edge devices such as the Arduino Nicla Sense ME. Additionally, they

gained insights into data collection and preprocessing from sensors, training and optimizing

models for low-energy environments, and applying these models to real-world scenarios in

agriculture, healthcare, and environmental monitoring.

Participants had the opportunity to interact with global thought leaders in the TinyML field.

Experts such as Dr. Marco Zennaro of ICTP, Professor John Shawe-Taylor from

IRCAI/UNESCO, Professor David Cuartielles from Arduino, Dr. Joan Pita Costa of IRCAI,

Professor Marcelo Ravai from the Federal University of Brazil, Thomas Basikolo from ITU

Geneva, and Malawi’s own Professor Chomora Mikeka shared their expertise. Their insights

emphasised how technology can drive sustainable development when grounded in local realities

and needs.