Hanselminutes is Fresh Air for Developers. A weekly commute-time podcast that promotes fresh technology and fresh voices. Talk and Tech for Developers, Life-long Learners, and Technologists.
…
continue reading
MP3•Episode home
Manage episode 513997254 series 2093893
Content provided by Kane Simms. All podcast content including episodes, graphics, and podcast descriptions are uploaded and provided directly by Kane Simms or their podcast platform partner. If you believe someone is using your copyrighted work without your permission, you can follow the process outlined here https://staging.podcastplayer.com/legal.
In this episode, we dive into the art and science of context engineering with Martin Cross, President of Connect, a company that empowers enterprises to transform through the use of AI and automation.
Martin shares how most businesses still approach AI with a technology-first mindset, deploying solutions before clearly defining the customer problem they’re solving. He explains why successful AI experiences begin with understanding the user journey, designing the right interaction, and then selecting the appropriate tools, not the other way around.
We also unpack Connect’s acquisition of Elerian AI and why contact centre voice recordings represent an untapped goldmine of training data. Martin explains how fine-tuning models with real-world conversations, including dialects, accents, and customer intents, can dramatically improve AI accuracy and empathy.
The conversation delves into the various layers of context - from session-level to journey-wide to CRM-integrated, and how they work together to transform generic interactions into deeply personalised, proactive experiences. We also reflect on why the industry continues to repeat old mistakes and how businesses can design AI that truly understands and anticipates customer needs.
Show notes
Follow Martin Cross on LinkedIn
Follow Kane Simms on LinkedIn
Take our updated AI Maturity Assessment
Subscribe to The AI Ultimatum Substack
Hosted on Acast. See acast.com/privacy for more information.
358 episodes