EPISODE · Mar 2, 2026 · 1H 8M
#03 of How Software Becomes Driving – Daniel Langkilde – When Autonomy Becomes a Data Problem
from BayInCo podden
How Software Becomes Driving is a BayInCo podcast series exploring how real-world AI systems are built, deployed, and scaled as vehicles transition from mechanical products into intelligent, software-defined and data-driven systems.In this episode, Kailash talks to Daniel Langkilde, Co-founder & CEO of Kognic, a technology leader working at the intersection of machine learning, real-world systems, and automotive autonomy. Together, they explore:• Why many autonomy challenges are information problems rather than modeling problems• The difference between what AI systems can express and what they can realistically learn• How human feedback underpins perception, labeling, and emerging multimodal systems• The impact of distribution shift and non-stationary environments on safety• The real role of simulation and synthetic data in closing, or hiding, learning gaps• How humans will remain essential teachers for machines as autonomy evolves Daniel brings a first-principles perspective on intelligence, uncertainty, and scale, highlighting why autonomy depends not just on data pipelines, but on understanding learning limits and designing systems that can operate responsibly in the real world.A must-listen for anyone working with autonomous systems, AI infrastructure, automotive software, or safety-critical machine learning.DisclaimerThe views and opinions expressed in this podcast are solely those of the individuals and do not reflect the views, policies, or positions of any companies or organizations they are affiliated with. This content is for informational and entertainment purposes only. Hosted on Acast. See acast.com/privacy for more information.
What this episode covers
How Software Becomes Driving is a BayInCo podcast series exploring how real-world AI systems are built, deployed, and scaled as vehicles transition from mechanical products into intelligent, software-defined and data-driven systems.In this episode, Kailash talks to Daniel Langkilde, Co-founder & CEO of Kognic, a technology leader working at the intersection of machine learning, real-world systems, and automotive autonomy. Together, they explore:• Why many autonomy challenges are information problems rather than modeling problems• The difference between what AI systems can express and what they can realistically learn• How human feedback underpins perception, labeling, and emerging multimodal systems• The impact of distribution shift and non-stationary environments on safety• The real role of simulation and synthetic data in closing, or hiding, learning gaps• How humans will remain essential teachers for machines as autonomy evolves Daniel brings a first-principles perspective on intelligence, uncertainty, and scale, highlighting why autonomy depends not just on data pipelines, but on understanding learning limits and designing systems that can operate responsibly in the real world.A must-listen for anyone working with autonomous systems, AI infrastructure, automotive software, or safety-critical machine learning.DisclaimerThe views and opinions expressed in this podcast are solely those of the individuals and do not reflect the views, policies, or positions of any companies or organizations they are affiliated with. This content is for informational and entertainment purposes only. Hosted on Acast. See acast.com/privacy for more information.
NOW PLAYING
#03 of How Software Becomes Driving – Daniel Langkilde – When Autonomy Becomes a Data Problem
No transcript for this episode yet
Similar Episodes
No similar episodes found.
Similar Podcasts
No similar podcasts found.