IRL & Online Program
Edge AI & IoT Deployment
Optimize and deploy lightweight neural networks onto microcontrollers, edge devices, and mobile hardware.
Hands-on Highlights
- Quantize and prune neural networks for edge devices
- Deploy models onto Raspberry Pi and Jetson Nano
- Build a wake-word detection system for microcontrollers
- Implement real-time inference on iOS/Android
Detailed Syllabus
Week 1-2
Model Optimization Techniques
- Why edge AI? Hardware constraints
- Post-training quantization (PTQ)
- Quantization-Aware Training (QAT)
- Network pruning and knowledge distillation
Week 3-4
Microcontrollers & TinyML
- Introduction to TinyML
- TensorFlow Lite for Microcontrollers
- Building a wake-word detection system
- Deploying on Arduino/ESP32
Week 5-6
SBCs & Hardware Accelerators
- Deploying on Raspberry Pi
- Accelerating inference with Google Coral TPU
- NVIDIA Jetson Nano basics
- Real-time object detection at the edge
Week 7-8
Mobile Deployment (iOS/Android)
- Converting models to CoreML (iOS)
- TFLite integration in Android Apps
- Handling battery/thermal constraints
- Final Edge AI capstone project
Target Roles & Career Paths
Edge AI Engineer
Embedded Machine Learning Engineer
IoT Developer
Mobile ML Engineer
These are the primary roles you will be equipped to apply for upon successful completion of the course and portfolio projects.