Hands on learning experience for non-technical people
summary
Claims that non-technical people can build a working AI robot from scratch and learn by doing, with a concrete pipeline: train virtual agents with genetic/reinforcement methods until competent, export the trained neural network, flash it into a low-power board (8‑bit, ~12 MHz) and run on a hand-assembled chassis with two motors and four light sensors. Training runs take a few hours, assembly about 30 minutes, parts cost around ~40 euros, and the physical robot behaves as expected but cannot continue learning.
AI-written summary · about
chapters (9)
- 0:00Introduction and background
- 3:23Fear of the unknown in tech
- 6:18Five keys to learning complex tech
- 11:53Hands-on learning with AI and robotics
- 22:02Building a physical robot
- 25:17Transferring AI to real world
- 27:58Learning by doing and experimenting
- 28:42Regaining connection to our hands
- 31:42Q&A and closing remarks
related talks
56mCan I do AI?
52mRead, Write, and Tinker with AI
10mMeet the Speakers
46mFullstack AI: The Building Blocks for Production Ready AI-Powered Software
25mNo Code, No Problem: Building Powerful AI Agents with Just Your Words
28mHands-on learning in the age of AI: are developer workshops still relevant
01h 05mQuick Start with AI for Businesses
09mSmale’s 18th Problem and Intelligence Science 2