Microcontrollers (MCUs), Cloud, and a Chance of AI with Tara Walker
summary
Claims that IoT jobs grew 1994% and that tiny MCUs can run real AI; demo builds and flashes an ESP32 with ESP-IDF using C/C++ and FreeRTOS, configures Wi‑Fi via Kconfig, produces bin/elf artifacts, and sends live temp/humidity/light readings to the cloud (Azure SDKs). Explains MCU vs MPU, shows Raspberry Pi contrast, and outlines using Tensorflow Lite for Microcontrollers to run tiny inference models on devices with ~KB–MB constraints; the demo hardware works and streams sensor data successfully.
AI-written summary · about
chapters (12)
- 0:00IoT and microcontrollers
- 3:52IoT operating systems
- 7:12Microcontrollers versus microprocessors
- 10:50Microcontroller anatomy
- 12:16Sensors and actuators
- 14:04Programming microcontrollers
- 15:32Real-time OS explained
- 17:16IoT cloud connection
- 18:18Live coding ESP32
- 24:10AI on microcontrollers
- 25:38Getting started resources
- 25:58Live sensor demo
related talks
01h 01mML and the IoT: Living on the Edge
24mRunning AI/Neural networks on microcontrollers made simple with the STM32Cube.AI
04mMultiple embedded and Cloud AI solutions at ST
19mWhy Machine Learning on Microcontrollers?
55mEdgeAI - AI in the Palm of Your Hand
59mML and the IoT: Living on the Edge
31mKeynote: The next big thing in machine learning is tiny
25mMachine Learning on Microcontrollers; Tiny, Useful and Fun