Doulos is pleased to announce the launch of the refreshed Essential Edge AI for Embedded Developers course - a significant update to our hands-on training program designed specifically for embedded engineers, system architects, and product managers.
This course provides a practical, end-to-end understanding of how to fine-tune and deploy AI/ML models to constrained edge devices, such as 64-bit Linux-based SBCs, microcontrollers, and neural network accelerators.
Built around a modern MLOps workflow, the course guides attendees through every stage of edge AI deployment. From foundational model development, data planning, and preprocessing, to model conversion, optimisation, deployment, and in-field monitoring.
In Essential Edge AI for Embedded Developers, we cover:
- MLOps workflow for edge AI applications
- Neural network basics including model training and Keras APIs
- Feature engineering and time series classification
- Model conversion between ONNX, TensorFlow Lite, and classical ML formats
- Porting trained models to microcontrollers using TinyML (TFLite Micro)
- Object Detection using SSD and YOLO models
- And much more...
The course does not focus on large-scale generative AI models or deep network architecture design (covered in the Practical Deep Learning course).
Instead, it provides actionable insight into deploying compact, efficient models for real-world embedded applications.
Who should attend?
Engineers involved in deploying AI/ML applications on constrained edge devices, including SBCs, 32-bit microcontrollers, and hardware accelerators.
Prerequisites:
Working knowledge of Python or C/C++, and a general understanding of embedded systems.
View the Edge AI for Embedded Developers course description and register your interest>
