Need autonomous driving training data? ›

New Podcast: The Challenges of Teaching Cars to Think Like Humans

New Podcast: The Challenges of Teaching Cars to Think Like Humans

Our friends at This Week in Machine Learning & AI interviewed our founder and CEO Daryn Nakhuda on their latest episode, “Training Data for Autonomous Vehicles.” (It’s the first in a new TWiML Talk series digging into the phenomenon of self-driving cars.)

You might remember our lead data scientist, Angie Hugeback, recorded a podcast with host Sam Charrington last fall about generating high-quality AI training data. A lot’s sure changed around here—their conversation didn’t even touch on all the hard work our customers in the automotive space are doing to make autonomy a reality. Today, delivering training data for autonomous vehicles in an efficient and safe way is Mighty AI’s top priority.

Daryn shared how our expertise in human insights and predictive machine learning across industries helped us evolve from Spare5 to a Top 10 Auto Startup.

Here are a few other topics Daryn and Sam discussed:

  • How concerns about scale will grow exponentially as more and more test fleets hit the roads
  • How autonomy has led to specialized requirements and an urgency for diverse global data
  • What role instructional design plays in accurately annotating terabytes of data
  • Why secrecy in the autonomous vehicle market won’t last
  • What ice cream trucks and slime eels have to do with teaching cars to understand their surroundings

Tune in below (or on iTunes, YouTubeSoundCloud, Google Play, or Stitcher):