The Agile Brand with Greg Kihlström®: Episode #785

Guest: David Funck, CTO, Avaya
Title: Building Persistent Memory of the Customer with AI
Date: December 17, 2025

Episode Overview

In this episode, host Greg Kihlström sits down with David Funck, Chief Technology Officer at Avaya, to discuss a pressing challenge in customer experience: how to break AI “amnesia” and create persistent, contextual customer journeys across platforms. Central to this conversation is the Model Context Protocol (MCP), a standard emerging to facilitate shared context between disparate AI systems—transforming how brands build customer lifetime value and operational agility, especially in enterprise contact centers. The discussion spans everything from AI sprawl and agentic AI, to tandem care between humans and AI, key metrics for success, and the practical rollout of interconnected, context-aware AI experiences.

Key Topics & Insights

1. Introduction to Persistent AI Memory & MCP

[01:07-05:31]

• The show opens with Greg asking: “What if every AI interaction with a customer built upon the last instead of starting from scratch every single time?”
• Challenge: AI features and “agentic AI” are everywhere, causing what Greg calls “AI sprawl”—“Everybody’s got their own AI features now...” [04:21]
• Problem: Fragmented customer journeys, where customers hop between isolated AI systems leading to inconsistent or repetitive experiences.

Model Context Protocol (MCP) Explained

[05:31-08:50]

• David Funck explains MCP:
  • “MCP stands for Model Context Protocol…a protocol that standardizes the way that large language models can interact with the real world.” [05:44]
  • Developed by Anthropic, MCP allows LLMs (Large Language Models) to access real-time information and take actions, rather than just predicting based on historical data.
  • MCP standardizes API integrations, enabling easier and more democratized connections across enterprise and web tools: “With MCP, you can pick a large language model…and connect them to…APIs…either in your enterprise back office or on the Internet at large.” [08:20]
  • Memorable quote:
    • “It’s a real game changer.” — David Funck [08:48]

2. Connecting Customer Journeys & Eliminating Fragmentation

[08:50-12:11]

• Legacy systems create “prescriptive” customer journeys—often irrelevant to actual customer needs.
• David Funck:
  • “You might be doing something totally different as a customer.” [09:52]
  • Agentic AI empowered by MCP enables adaptive, dialog-driven experiences responsive to each customer—not locked into rigid flows.
• Avaya’s Approach:
  • Their solution, Avaya Infinity, enables experimentation by first empowering human agents, then scaling to end customers once guardrails are tested.
  • “AI is perfectly suited to handling those…routine things. And then if you can help a human agent…it leaves time for that human agent to do things that only a human can do…” [11:17]
  • Vision: A seamless partnership, or “tandem care,” where AI and humans complement each other.

3. Elevating Human Roles & Strategic AI Partnership

[12:11-14:19]

• Greg: “AI could…form a partnership with the humans to also augment some of that higher thinking?”
• David Funck:
  • “That’s definitely not overly optimistic…We’re talking about this idea of tandem care, right? Of AI and humans working together to improve the care that customers get…” [13:06]
  • Tandem care starts with AI handling routine, repetitive tasks, freeing humans for complex, empathic interactions—but also hints at AI helping with higher-order tasks alongside humans, especially in knowledge worker scenarios.

4. Measuring AI’s Impact: Beyond Traditional Metrics

[16:42-19:17]

• Traditional metrics (e.g., average handle time, first contact resolution) are still important.
• David Funck:
  • “All of the capabilities that we use to measure the human agent, we’re also using our same analytics package to measure the effectiveness of AI.” [17:31]
  • AI’s effectiveness and return on investment (ROI) become trackable and comparable against human agents.
  • Enterprises can train more focused, domain-specific LLMs which are “much more cost effective to run.” [18:37]
  • Cost effectiveness, accuracy, and domain relevancy are the new frontier in contact center metrics.

5. MCP Rollout & Broader Enterprise Potential

[19:18-20:15]

• Avaya’s MCP Roadmap:
  • “We’re demoing it right now…It’ll be available in pilot at the end of Q1 and then…it’ll be available in production…in calendar Q2 of this year of 2026.” [19:33]
• Beyond Contact Centers:
  • Knowledge workers benefit from LLMs able to connect multiple back office tools (code repos, ticketing, documentation) via MCP.
  • David Funck:
    • “It really is a wide open horizon…what we are going to do as a society with that additional productivity. Gets back to this tandem care idea…” [20:56]

6. The Future of AI in Enterprise: Specialization Over Scale

[21:59-23:43]

• Prediction for the Next Year:
  • “Specific models getting really, really good with very focused constraints so that they don’t hallucinate, getting very, very good at specific tasks. …it’s much more cost effective because the models are smaller.” [22:21]
  • The future is many specialized, highly effective models serving defined niches—shifting away from just massive general-purpose models.

Notable Quotes & Timestamps

• On AI Sprawl:

  “Everybody’s got their own AI features now. Most have their own agentic component. So we’re kind of at this place…what do you choose? …The ones who suffer the most…are the customers.”
  — Greg Kihlström [04:21]

• Defining MCP:

  “MCP is a protocol that standardizes the way that large language models can interact with the real world.”
  — David Funck [05:44]

• MCP as a Game Changer:

  “It gives the model the ability to reach out and ask a question about the real world or take action…”
  — David Funck [07:23]

• Agentic AI & Customer Journeys:

  “It becomes then a dialogue…with the artificial intelligence. …if you can help a human agent with those routine things…then it leaves time for that human agent to do things that only a human can do…”
  — David Funck [10:00, 11:17]

• On Tandem Care:

  “We’re talking about this idea of tandem care, right? Of AI and humans working together to improve the care that customers get in the contact center.”
  — David Funck [13:06]

• Enterprise Application of MCP:

  “It really is a wide open horizon …what we are going to do as a society with that additional productivity. Gets back to this tandem care idea…”
  — David Funck [20:56]

• What’s Next for AI Models:

  “Specific models getting really, really good with very focused constraints so that they don’t hallucinate…we’ll see more and more of those kinds of things happening in the next year.”
  — David Funck [22:21]

• Leadership & Staying Agile:

  “I like to surround myself with really cool and effective people…if we’re miserable at work, we’re wasting a lot of our life. I try to make every meeting as fun as possible and that keeps me on my toes…”
  — David Funck [24:00]

Timestamps for Important Segments

| Timestamp | Segment Description | |--------------|---------------------------------------------------------------------| | 01:07–05:31 | Introducing MCP, AI sprawl, and customer pain points | | 05:31–08:50 | Technical breakdown of MCP and its implications for AI | | 08:50–12:11 | How MCP and agentic AI improve customer journeys and contact centers | | 12:11–14:19 | Human-AI partnership (“tandem care”) and elevating agent roles | | 16:42–19:17 | Measurement, analytics, and ROI in AI-powered contact centers | | 19:18–20:15 | Avaya’s MCP rollout and demonstration timeline | | 20:15–21:59 | Broader enterprise applications: knowledge worker productivity | | 21:59–23:43 | Predictions for the next year: focus on specialized AI models | | 23:53–24:40 | David Funck shares agile leadership philosophy |

Takeaways

• Shared AI memory through MCP can unlock seamless, context-rich customer journeys.
• Agentic AI, guided by standardized protocols, bridges silos and eliminates repetitive "phone tree" doom loops for customers.
• The future of enterprise AI is a symbiosis: human empathy and oversight paired with AI’s efficiency—what Avaya calls “tandem care.”
• Metrics and ROI remain critical—AI’s cost-effectiveness and performance will become as measurable as human agents’.
• The horizon is wide open: MCP’s biggest promise may be unlocking new productivity and value across all knowledge work, not just contact centers.