Explorez tous les épisodes du podcast Deep Learning with PolyAI

An agent that gets better every day compounds into a very different agent within months. That's the idea behind Wren, PolyAI's new release: an agent whose job is to make your other agents better. Some of what it finds is a small refinement, some of it is a big fix. Either way, Wren watches every real customer conversation, learns from what happens, and keeps improving your customer-facing agents.

 Nikola Mrkšić sits down with Arkadiusz Kwapiszewski, Head of Agent OS, to get into what that changes, how Wren is putting agent-building in the hands of people who could never touch it before, and why 95% of PolyAI's own production changes now ship through it. 

Listen to the full episode, and learn more about PolyAI at https://poly.ai?utm_source=youtube&utm_medium=podcast&utm_campaign=podcast&utm_content=podcast

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Companies are pouring money into AI, but not everyone is getting the returns. Ian Jacobs, VP and Lead Analyst at Opus Research, thinks he knows why. In this episode, your host Nikola Mrkšić sits down with Ian to work through two ideas from his latest research. 

The first is Conversation Experience Orchestration: the wall between conversational AI and conversation intelligence is coming down, and conversations are becoming the layer that decides what happens next. 

The second is the AI agent control plane: as enterprises get pitched agents by every vendor at once, they need a shared operating layer to keep them coordinated. 

Listen to the full episode, and learn more about dialog AI at https://poly.ai?utm_source=youtube&utm_medium=podcast&utm_campaign=podcast&utm_content=podcast

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

When PolyAI built its Builder MCP, one client helped shape it: Vixxo Facility Solutions, one of the largest facilities providers in the US. In this episode, guest host Damian Sasso talks with Vixxo's CTO Derek Neighbors and VP of IT Bobby Hunnicut about getting AI agents into the hands of people who've never written code, from accounting to technicians in the field. They cover why an MCP or CLI has become their litmus test for adopting any technology, how they build and ship their own MCPs as they go, why REST APIs are still the foundation, and how to stop governance debates from swallowing the whole conversation. 

See how PolyAI's Builder MCP works at https://poly.ai?utm_source=youtube&utm_medium=podcast&utm_campaign=podcast&utm_content=podcast

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

PolyAI just launched Dialog-RSN-1, its first audio-native model, and your host Nikola Mrkšić sat down with its builder, Matt Henderson, to unpack why it’s a game-changer for building voice agents.

Most voice AI either flattens a call into a transcript and loses the audio, or goes fully speech-to-speech and gives up control of the voice. Dialog-RSN-1 does neither. It hears the raw audio directly, decides when to speak, and keeps text-to-speech separate so the voice stays under your control, all in under 300 milliseconds.

Nikola and Matt get into what audio-native really means, how auto-reasoning keeps it fast, and why it beats every other real-time model on quality and quickness. Hear the full episode, and see how PolyAI builds dialog agents that hear the whole call at https://poly.ai?utm_source=youtube&utm_medium=podcast&utm_campaign=podcast&utm_content=podcast

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Voice AI vendors love to quote word error rate, usually somewhere around 2 to 3%, as proof their system understands customers. Oliver Shoulson, Agent Design & Engineering Lead at PolyAI, thinks it's one of the most misleading numbers in the business. He and Nikola Mrkšić go through a month of analysis across more than 100,000 real production call turns to explain why. 

Turns out not all words carry the same weight, most transcription errors are harmless, and task success barely moves even when the system mishears, because today's models recover the way a person would. 

Listen to the full episode, and see how PolyAI builds dialog agents that get this right at https://poly.ai?utm_source=youtube&utm_medium=podcast&utm_campaign=podcast&utm_content=podcast

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Engineering teams are shrinking. But is that a crisis or an upgrade?

In this episode of Deep Learning with PolyAI, Nikola Mrkšić sits down with Helen Greul, PolyAI's new SVP of Engineering, to explore what AI is doing to the shape of technical teams, the nature of engineering leadership, and who actually thrives when the rules change.

They discuss why a new generation of non-traditional builders is outpacing seasoned development teams, how smaller nimble units with shared context are replacing large headcounts, and why complexity doesn't disappear when AI arrives: it just moves somewhere else.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Ten minutes to a working AI agent, tested and ready to deploy.

In this episode of Deep Learning with PolyAI, Frank Ferro sits down with Ruari Phipps, Senior Forward Deployed AI Engineer at PolyAI, for a live demo of the Agent Development Kit — and a look at what it actually means to give developers real control over how they build AI agents.

They explore why enterprise AI platforms so often force a tradeoff between speed and control, how the ADK changes that equation, and what it looks like when a developer pairs the toolkit with an AI coding agent like Claude Code to go from a blank canvas to a fully tested, branch-ready agent in real time.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

What happens when the ability to build an AI agent is no longer limited to developers?

In this episode of Deep Learning with PolyAI, Nikola Mrkšić sits down with Michael Chen, VP of Strategic Alliances at PolyAI, to explore what's driving the explosion in AI agent development — and what it means for how enterprises build, deploy, and think about AI.

They cover the rise of personal AI agents, why Jensen Huang's $1 trillion data center bet might 

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Many AI models today are designed as generalists. But in customer experience, that approach often falls short.

In this episode of Deep Learning with PolyAI, Nikola Mrkšić sits down with Matt Henderson, VP of Research at PolyAI, to unpack why specialized AI models outperform general-purpose ones in real-world CX environments.

They explore the tradeoffs between speed, accuracy, and reasoning, why voice AI requires fundamentally different design decisions, and what makes PolyAI’s Raven 3.5 a step forward. The conversation offers a behind-the-scenes look at what it actually takes to build AI that works in production, not just in demos.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

In this episode of Deep Learning with PolyAI, Nikola Mrkšić sits down with Bill Staikos, Founder of Be Customer Led, to unpack why many CX leaders are still on the sidelines of the AI conversation — and what that means for the future of the function.

They explore how AI is reshaping the “interaction layer” of customer experience, why traditional CX metrics like NPS are no longer enough, and how leading organizations are starting to tie CX directly to revenue, cost, and risk.

The conversation challenges CX leaders to think bigger and make their function into an outcome-driven engine for business growth. In an AI-driven world, the leaders who understand that shift will have a much bigger role to play.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Customer experience is often measured at the surface. But it’s built behind the scenes.

In this episode of Deep Learning with PolyAI, Michelle Schroeder sits down with Sam Stern, who leads service design at LinkedIn, to explore how employee experience shapes every customer interaction.

They discuss how LinkedIn is improving CX by fixing the systems employees rely on — from reducing “toggle tax” across tools to giving teams better data and real-time guidance with AI. The result is not just faster service, but more meaningful, higher-quality interactions when it matters most.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

In this episode of Deep Learning with PolyAI, Michelle Schroeder sits down with Nate Brown, Head of Education and Enablement at Metric Sherpa and cofounder of CX Accelerator, to discuss how AI is reshaping the way companies think about customer experience.

Nate shares a framework that’s gaining traction with CX leaders: make the quick parts quick, and make the slow parts meaningful. Automation should remove friction and resolve routine problems quickly, but the moments that matter most still require guidance, empathy, and human connection.

Together they explore how organizations can use AI to reduce effort for customers while still creating experiences that build loyalty, trust, and long-term value.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

In this episode of Deep Learning with PolyAI, Nikola Mrkšić sits down with Steven Fine, GM of Restaurants at PolyAI, to explore how technology has shaped the customer experience in restaurants over the past century.

From the first phone reservations to online booking platforms and now AI-powered voice agents, the conversation traces how hospitality has continuously adapted to meet changing customer expectations.

Nikola and Steven discuss why the phone is still one of the most important touchpoints in the restaurant journey, how missed calls translate into lost revenue, and how AI is helping restaurants capture demand while improving both staff and guest experience. Along the way, they explore no-shows, loyalty programs, large party bookings, and why voice AI is becoming a new revenue engine for the hospitality industry.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

In this episode of Deep Learning with PolyAI, Michelle Schroeder sits down with Sarah Wright, Vice President of Global Customer Service at ResMed, to talk about what it really takes to transform healthcare CX. Sarah describes the challenge as an Everest climb. Every step forward requires balancing innovation with responsibility, especially when patient health information is involved.

Unlike other industries, healthcare cannot experiment loosely. Data must be protected. Integrations must be intentional. The order of operations matters.

• In this conversation, they discuss:
  • Why protecting patient health information shapes every AI decision in healthcare
  • How fragmented systems make personalization difficult
  • What it takes to move from reactive break-fix service to predictive, personalized experiences
  • How “data stories” help build internal alignment for transformation
  • Why voice of customer and QA datasets are often the most powerful starting point
  • How AI should augment human moments rather than eliminate them
  • Why frontline teams become brand defenders and trust gatekeepers as automation scales
  • The importance of starting with baselines and defining what “good” actually looks like

Sarah also shares practical advice for leaders under pressure to modernize quickly: start with the friction that shows up in your data, focus on high-volume pain points, and avoid trying to fix experiences that patients already value.

Climbing healthcare’s AI Everest takes patience, humility, and strong change leadership. This episode offers a realistic look at what that journey involves.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

In this episode of Deep Learning with PolyAI, your host Nikola Mrkšić chats with Alex Brown to unpack what happens when AI takes over the administrative work that quietly defines modern healthcare. Instead of replacing doctors and clinicians, AI is reshaping everything around them:  from documentation and scheduling to the workflows that consume hours of time.

The conversation explores why healthcare resisted automation longer than other industries, why AI’s fastest wins are operational rather than clinical, and how removing administrative friction could improve experiences — for both providers and patients.

Tune in to hear:

• Why healthcare’s biggest AI gains are operational, not clinical
• How admin-heavy workflows shape both provider burnout and patient experience
• Why “AI replacing doctors” misses the real opportunity
• What efficiency actually looks like in a healthcare context
• Where empathy still matters — and where automation clearly helps

If you’re curious about where AI is quietly reshaping healthcare today, tune in, and let us know what you think.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

In this episode of Deep Learning with PolyAI, Nikola Mrkšić sits down with Nathan Pearson, GM, Financial Services at PolyAi, to trace the evolution of personalized banking. From branch tellers and ATMs to digital channels and AI-driven experiences, they explore how banks have historically balanced scale, efficiency, and trust, and why personalization has always been central to customer relationships.

The conversation looks at why personalization has always mattered in financial services, how each new channel changed customer expectations, and what today’s AI moment has in common with past shifts. While technology keeps changing, the core challenge for banks hasn’t: delivering personal, trusted service at scale.

Together, they discuss:

• How early branch banking set the original standard for personalized service
• How online and mobile banking reshaped customer expectations around convenience
• Why “self-service” has never meant “impersonal” for customers
• How each new technology promised efficiency, but introduced new tradeoffs
• What banks can learn from history as they deploy AI across customer journeys

The big takeaway: while tools keep changing, the goal hasn’t. Customers still want banking experiences that feel personal, reliable, and easy — even when they’re powered by automation.

If you’re thinking about how AI should support (not replace) personalization for your business, be sure to tune in.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

AI has become remarkably good at answering questions — but customer experience should go beyond answers. It’s about resolution. Too often, AI fails when customers can’t easily explain what they’re seeing on their screen.

In this episode of Deep Learning with PolyAI, Jenn Cunningham sits down with Zaccari Scalzi of Cobrowse.io to explore why visual context is becoming essential for effective customer service AI. Together, they discuss how combining conversational AI with real-time visual grounding helps agents — human and AI — understand complex, multi-step customer journeys.

They cover:

• Why traditional bots break down in real customer interactions
• How visual context enables faster, more accurate resolution
• The difference between containment and true problem-solving
• How AI-to-human handoffs improve when everyone sees the same thing
• Why governance, consent, and privacy matter when AI “sees” the screen

The takeaway is simple: AI works best when it has full context. And for many customer journeys, that context is visual.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

In this episode of Deep Learning with PolyAI, Michelle Schroeder sits down with Dom Nicastro, Editor-in-Chief at Simpler Media Group and longtime journalist covering CX, content, and enterprise tech.

Dom has spent years watching AI reshape newsrooms — and he sees striking parallels with what CX leaders are facing today. Journalists and CX teams share the same challenge: cutting through hype, spotting “AI slop,” and building trust through accuracy, governance, and human judgment.

They explore:

• Why journalism’s skepticism is exactly what CX needs right now
• How AI accelerates good work — and exposes weak processes
• Why bad AI (in news or CX) destroys trust instantly
• What Dom hears backstage at CX events that rarely makes it into press releases
• The role of governance, vetting, and human oversight in customer-facing AI
• How CMSWire built “Roberta,” their internal RAG system to support real reporting — not replace it

Dom’s take is clear: AI won’t replace humans, but it will reveal who’s doing the work well — and who isn’t.

👉 Listen to the full episode to learn what journalism can teach CX leaders about building trustworthy AI.

#AI #CustomerExperience #CX #Journalism #PolyAI #CMSWire #ConversationalAI

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Every customer expresses themselves differently, and their choice of language is no exception. So how can brands bridge the gap when customers come to them speaking different languages?

In this episode of Deep Learning with PolyAI, Jenn Cunningham sits down with Matt Henderson, VP of Research, and Viola Lin, Product Manager, to explore the evolution of multilingual voice AI and what it means for global customer experience.

They unpack what “multilingual” really means in practice — from accurate translation to cultural intelligence — taking PolyAI’s AI models and Agent Studio platform as a starting point to discuss helping enterprises deploy agents that sound natural, respectful, and consistent in over 45 languages.

They explore:

• Why multilingual CX should factor in much more than just translation, including cultural nuances
• How PolyAI handles dozens of languages in a single deployment
• Real-world challenges and what to do about them: from accents and formality to tone and gendered language
• The role of fine-tuning, voice selection, and design in making AI sound human
• How global brands use multilingual AI to deliver support that feels "local" everywhere

👉 Learn how multilingual AI is bringing accessibility and empathy to CX.

#AI #VoiceAI #CustomerExperience #PolyAI #MultilingualAI #ConversationalAI #CX

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

At VOX 2025, one thing was clear: the best brands aren’t just adopting AI — they’re becoming fluent in it.

In this episode of Deep Learning with PolyAI, Michelle Schroeder, CMO of PolyAI, sits down with Christopher Osborne, PolyAI’s Chief Product Officer, to unpack what we learned from our annual conference about how the world's top enterprises are moving from early adoption to true fluency in AI for customer experience.

They reflect on key lessons from VOX 2025, where leaders across hospitality, banking, healthcare, and more shared how they’re using agentic AI to scale exceptional service without losing the factors that make them stand out to their customers.

They discuss:

• How Seminole Hard Rock, Boyd Gaming, and Fertitta Entertainment use AI to protect the guest experience, not replace it
• How Metro Bank is turning “a bank in every pocket” into a banker in every pocket
• How DOCS Dermatology, Simplyhealth, and SafeRide Health are expanding access to care through AI-driven patient engagement
• Why the next CX differentiator isn’t adoption — it’s fluency
• How PolyAI helps enterprises balance automation with humanity

👉 Watch the full episode to see what AI fluency really looks like inside the world’s most customer-obsessed brands, and watch the full VOX conference here.

#AI #CustomerExperience #PolyAI #VOX2025 #AgenticAI #VoiceAI #AICX

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

The conversational AI market has exploded: ~650 companies, even more tools, and way too many promises. For business leaders, navigating all this can be paralyzing.

In this episode of Deep Learning with PolyAI, your host Nikola Mrkšić sits down with Dan Miller, founder of Opus Research and the analyst who coined the term "conversational commerce," to unpack why the AI ecosystem feels more crowded and confusing than ever.

Dan calls it the “Deer in Headlights” moment: a point where too many choices, too much hype, and too little clarity leave enterprises frozen on the road to AI adoption.

They discuss:

• Why the AI boom has created confusion instead of clarity
• The Botenfreude phenomenon — why people still enjoy watching bots fail
• How most AI “failures” come from poor design, not bad tech
• What it will take for AI vendors to cooperate instead of compete
• Why success now depends on defining clear goals and working together

Dan has been tracking and shaping this space for decades. Watch the full episode to learn how we can move past the “deer in headlights” moment and start driving real progress.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Hello Sugar is one of the fastest-growing service brands in America, with more than 140 locations and counting. From the start, they built a text-first, digital experience that customers loved. But growth brought new insights: even in a digital-first world, voice calls were still critical to winning — and keeping — customers.

In this episode of Deep Learning with PolyAI, Nikola Mrkšić talks with Austin Towns, CTO of Hello Sugar, about how the brand rethought customer experience at scale. With voice AI agents, Hello Sugar turned a challenge into a growth advantage.

We discuss:
- Why even digital-first customers still pick up the phone
- How Hello Sugar blends automation with human touch
- The role of AI in scaling service without scaling costs
- Why voice remains a powerful channel for hyper-growth brands

👉 Listen to the full episode to learn how Hello Sugar is setting the standard for CX in high-growth service businesses.

#CustomerExperience #CX #AI #VoiceAI #PolyAI #HelloSugar #HyperGrowth

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Large language models are powerful tools — but why do they ramble on and on instead of just getting to the point?

In this episode of Deep Learning with PolyAI, Nikola Mrkšić sits down with Oliver Shoulson, one of PolyAI’s original dialogue designers, to unpack why LLMs talk the way they do, and what that means for building natural-sounding voice AI.

They dive into:

• Why LLMs are trained to over-explain, and how that breaks spoken dialogue
• The subtle design choices that can make conversations feel human — without trying to fool anyone
• The debate over anthropomorphism: should AI agents admit they’re AI, or lean into human-like traits?
• How in-house LLMs can be fine-tuned to sound less like overeager interns and more like trustworthy teammates

Voice AI doesn’t need to trick us into thinking it’s human, but it does need to feel usable. This episode explores the hidden details of dialogue design that decide whether AI earns customer trust, or just frustrates them with endless rambling.

👉 Watch the full conversation and learn more in Oliver's blog post about how to prevent your customer conversations from sounding like they're LLM slop: https://poly.ai/blog/how-not-to-talk-like-an-llm/?utm_source=youtube&utm_medium=podcast&utm_campaign=agenticai&utm_content=podcast

#VoiceAI #LLM #Anthropomorphism #CustomerExperience #PolyAI

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Most companies measure CX through surveys, reviews, and dashboards. But there is real danger in “silent complaints”: the underwhelming experiences that customers never report, but quietly share with friends or that stop them from coming back.

In this episode of Deep Learning with PolyAI, your host Nikola Mrkšić sits down with Adrian Swinscoe — author, Forbes columnist, and host of Punk CX — to explore how AI is reshaping customer experience, and why technology alone won’t save you from these hidden risks.

They discuss:

• Why “silent complaints” are often more damaging than bad reviews
• How AI can surface insights from conversations that surveys and dashboards miss
• Why frontline employees still see problems before the data does
• How leaders can balance technology with humility, curiosity, and human insight
• What it really takes to deliver on the old promise of being “customer-obsessed”

AI is giving companies new tools to understand customers in real time, but unless it’s used to address the basics of service, it risks becoming just another layer of CX hype.

👉 Watch the full episode to learn how to spot silent and not-so-silent complaints before they destroy loyalty to your brand.

#CustomerExperience #CX #AI #VoiceAI #PolyAI #PunkCX

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Analytics has always been the hardest part of the contact center. Dashboards are slow, incomplete, and rarely used outside of specialist teams. But what if analytics could be as simple as asking a question?

In this episode of Deep Learning with PolyAI, your host Nikola Mrkšić brings back Damian Sasso to explore how agentic AI is transforming analysis itself. Instead of waiting weeks for trends to surface, leaders can now interact with their data in real time: surfacing insights, testing hypotheses, and finding the “needle in a haystack” moments that drive real business change. Capabilities like PolyAI's new Smart Analyst are making it all come to life, and Damian shares a demo to show exactly how it all works.

The conversation touches on:
- Why analytics has lagged behind in AI transformation
- How unstructured conversation data is finally becoming accessible
- The shift from static dashboards to dynamic dialogue with data
- What this means for the evolution of the contact center — from a cost center to a true command center

Watch this episode for a glimpse into how agentic AI teams will help leaders see, understand, and act on their customer conversations. And learn more about Smart Analyst at: https://poly.ai/blog/polyai-smart-analyst/?utm_source=youtube&utm_medium=podcast&utm_campaign=agentstudio&utm_content=podcast

#AI #VoiceAI #CustomerExperience #ContactCenters #PolyAI

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

We all know AI agents can answer calls (with varying success rates depending upon their sophistication). But what if AI agents could also help CX leaders analyze their deployments, ask questions about their unique challenges and opportunities, and continuously improve?

In this episode of Deep Learning with PolyAI, Nikola Mrkšić sits down with Damian Sasso, Group PM at PolyAI, to explore how the field is moving from solely customer-facing AI agents toward agentic AI teams that can do all of the above.

We cover:

• Why MIT says 95% of AI pilots fail — and how to be in the successful 5%
• The role of QA Agents in monitoring performance
• How Analyst Agents like Smart Analyst surface business insights in real time
• The promise of Builder Agents that drive continuous optimization
• Why contact centers are shifting from call centers to command centers

If you want to understand how agentic AI is changing the landscape — one in which AI agents don’t just talk, but think, analyze, and build — this episode is for you.

👉 Subscribe today for more info about AI for CX.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

🚨 OpenAI just made its Realtime API generally available, and it's got everyone talking. But is it a true game-changer, or just another update?

In this episode, your host Nikola Mrkšić (Co-founder & CEO, PolyAI) is joined by Shawn Wen (Co-founder and CTO, PolyAI) to break down what this release really means:

🔹 Why Realtime API could kill the old turn-taking model
🔹 How speech-to-speech changes latency, interruptibility, and naturalness
🔹 What benchmarks actually say about performance
🔹 The impact on contact centers, CX leaders, and startups building voice agents

Whether you’re a CX leader, product builder, or just curious about the latest in voice, check out Nikola and Shawn's conversation to learn more about the news.

#OpenAI #RealtimeAPI #VoiceAI #AI #CustomerExperience #PolyAI #ContactCenters

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Ever wonder what really makes an AI voice feel human? In this episode of Deep Learning with PolyAI, an all-star panel from our dialogue design and engineering teams takes you behind-the-scenes to share how the hardest part of CX automation actually works.

From latency and personalization to multichannel design and our in-house Raven model, they break down the science (and the art) behind building conversational AI agents that are actually worth talking to.

You’ll learn:
✅ Why dialogue design is notoriously hard, and how PolyAI cracked it
✅ The key factors that make conversations “buttery smooth”
✅ How omnichannel and LLM-powered design are changing everything
✅ What’s next for dialogue design, from multimodal to configurable latency

PolyAI's Jenn Cunningham guest hosts for this episode, moderating a discussion between Arkadiusz Kwapiszewski, Eshan Singhal, and Harry Swanson.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

In this episode, your host Nikola Mrkšić sits down with Matt Henderson, VP of Research at PolyAI, to unpack the state of large language models, their quirks, and what really matters when building smarter voice agents for customer service.

Join us for a discussion on:
- Why GPT-5 may feel more incremental than game-changing
- Why reasoning models can fail at surprisingly simple tasks
- How PolyAI's Raven outperforms generalist LLMs in latency-sensitive, real-world CX use cases
- The balance between speed, accuracy, and reasoning for live customer interactions
- What open-source models, quantization, and fine-tuning mean for enterprise AI strategies

👉 Learn more about Raven and conversational AI here: https://poly.ai/blog/polyai-raven-v2-large-language-model/

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Brian Jeppesen, who serves as Director, Contact Center Operations at Fertitta Entertainment, joined your host Nikola Mrkšić to discuss his vision for the AI-powered contact center.

Listen to hear Nikola and Brian cover:

• How to overcome the challenges of staffing call centers during high-demand periods, including with virtual agents
• Their reaction to the Keep Call Centers in America Act of 2025, and what it means for brands and BPOs 
• Myths around the implementation of AI in contact centers 
• The risks BPOs (Business Process Outsourcers) face if they do not proactively offer AI solutions to their clients
• The transition to generative AI-powered virtual assistants, including what's working today and what's still yet to come.

PolyAI is proud to bring insights from our customers to the pod! Learn more about our work together here: https://poly.ai/case-studies/golden-nugget/

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

PolyAI co-founder and CTO Shawn Wen joins Nikola for a deep dive into the just-released OpenAI GPT-5 model. From model-flipping and 4o's depreciation to curbing sycophancy and determining which LLM out there is #1 for coding, they cover it all. Join us for a discussion on:

• OpenAI's bold approach to upgrading and removing models 2
• Hands-on experiences with GPT-5 and its capabilities for coding and other tasks
• Trends in AI, including the increasing use of language models as digital concierges 
• Concerns about transparency in benchmarks

GPT-5 is now available in PolyAI's Agent Studio. Learn more at https://poly.ai/

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

This week, Nikola is joined by Becky Kershaw, Deployments Lead at PolyAI.

Becky shares how using her own British accent for AI voice cloning increased engagement in outbound sales, showing the power of personalization and cultural fit.

They also discuss the crucial role of forward-deployed engineers—AI engineers, product managers, and dialogue designers—who immerse themselves in the customer’s world to deliver real results by:

- Solving complex problems
- Integrating with fractured legacy systems
- Continuously fine-tuning AI based on real-world feedback

Finally, Becky highlights the importance of linguistic expertise in shaping voices and prompts that make AI experiences effective and user-friendly.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Following up from Nikola Mrkšić and Nathan Benaich's recent article for Fortune -- ‘AI rollup’ investors think services firms can trade more like software companies. Here’s what they get wrong. Read here -- https://fortune.com/2025/06/27/ai-rollup-investment-strategy/

The buzz around AI roll-ups - buying low-margin service businesses and turning them into high-margin software companies - sounds enticing. But it's not that simple.

While AI can enhance efficiency, it doesn't automatically transform business models. Many service industries still value human expertise and trust. The real challenge lies in balancing AI integration with maintaining customer relationships and service quality.

As AI adoption accelerates, how will your industry adapt? Will AI complement or replace human touch in your field?

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Fortune article: http://fortune.com/2025/06/27/ai-rollup-investment-strategy/

Nikola reflects on his recent Fortune commentary (above) into the transformative potential of AI rollup strategies in Business Process Outsourcing (BPO). He and Damien discuss the concept of AI rollups (where private equity firms consolidate service companies to create more efficient, tech-enabled businesses), as well as  the unique challenges and opportunities associated with integrating AI into existing BPOs, the role of Global Systems Integrators (GSIs), and the evolving landscape of customer service automation. Hear why the world won't be rewritten by AI overnight but will transform progressively" in this week's Deep Learning!

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Nikola and Damien are back to discuss the ongoing evolution of AI in the contact center industry. We'll explore whether AI is truly replacing human workers or simply augmenting their capabilities, covering Klarna's automation efforts and recent Gartner predictions, and analyzing the challenges and opportunities in implementing AI solutions. We touch on essential aspects of project management, customer service strategies, and the future of AI integration in contact centers. Join us, or be assimilated!

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

VP of Product Marketing Brian Thompson joins Nikola to discuss the current state and future prospects of AI in financial services, including banking and insurance. The chat covers the importance of human oversight in AI deployments, the evolving landscape in banking and FinTech, optimization of back-office processes, and how financial institutions are leveraging generative AI. Brian shares insights on the differences between large legacy banks and smaller community banks, and the challenges and opportunities AI presents in bridging these gaps, while tackling customer expectations, regulatory concerns, and the significance of a strategic approach to AI implementation.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Damien and Nikola are back to discuss the transformative impact of generative AI in our everyday lives. The conversation covers everything from the rise of LLMs in personal development and therapy to DIY projects and coding. Discover how generative AI is reshaping industries, solving everyday problems, and why it's gaining traction in fields like legal services and medical advice. as well as future applications of AI.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Guest host Meghan Berton and VP of Engineering Razvan Kusztos cover the release of PolyAI's Owl ASR model, while dishing goss on the intricacies of developing high-performing speech recognition models. They discuss the significance of in-house SLU tech, the challenges of ensuring accuracy across different accents and environments, and the benefits of using synthetic data. The conversation also covers the cultural nuances in text to speech models and how Poly AI leverages partnerships with companies like NVIDIA for enhanced model performance. 

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Damien and Nikola discuss the brave initiatives taken by executives in the field of generative AI, their successes, and challenges. Nikola Mrkšić, CEO and co-founder of PolyAI, shares insights on differentiation in the generative AI space, the viability of build vs. buy decisions, and the importance of proof of concept in AI projects. The episode sheds light on the technological advancements and strategic decisions that help organizations capitalize on generative AI, with illustrative case studies and industry comparisons.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Damien and Nikola discuss the implications of AI-first approaches in industries, with a focus on how companies like Duolingo and Klarna are navigating the AI transformation. We explore the concept of an intelligent enterprise operating system and debate whether traditional companies can adapt quickly enough. Additionally, we examine the potential impacts on human employment and strategies for businesses to keep their employees motivated amidst these changes.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

In this buzzy episode, Damien hosts PolyAI's Senior Vice President of Product, Devidas Desai, to announce the launch of Agent Studio, the world's only voice-first omnichannel CX platform for AI agents. Devidas provides an in-depth look into what Agent Studio is and discusses how the platform has been rigorously tested with live enterprise deployments and includes proprietary technologies like advanced automatic speech recognition and specialized language models. Hear about flagship tech like 'Talk Tracks,' and more of its future enhancements.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Damien and Nikola are back to discuss the impact of removing toll-free numbers for government services, focusing on the DOGE/Social Security Administration's briefly considered decision. They explore the potential operational and cost-saving implications of voice AI in government services, the challenges of optimizing such large systems, and the long-term strategies for digital transformation. The discussion also touches on the changing preferences of Generation Z towards voice communication and the importance of careful, well-planned implementation of automation in government operations. 

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Nikola and Damien chew the fat about developments from AI giants Anthropic and OpenAI, comparing their new models: Claude 3.7 and GPT-4.5 Orion, and where they place in the consumer and enterprise markets. We also touch on OAI's specialized agents for hire, Amazon's new agentic AI division and the recent stance of the US Patent Office on AI-generated content.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

In the closing episode of Deep Learning's deep dive on telephony, Nikola and Oisin discuss innovative ideas in telephony, especially the concept of using a universal AI ringtone to replace lengthy warnings and terms during calls. They debate the efficacy of this idea and its potential global impact on user experience. They also touch on the challenges of integrating AI within legacy telephony systems and explore different modern communication protocols like Gibberlink. The conversation also highlights the inefficiencies of current hold systems and the untapped potential of AI in streamlining B2B interactions.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

The second episode of a three-part miniseries sees Deep Learning with PolyAI continue exploring the intricate world of telephony and VoIP! Nikola welcomes back Oisin Glenn, VP of Solutions Consulting, to chat about the historical development, current technologies, and the complexities of modern telephony. They discuss everything from PBX systems, VoIP protocols, and the rise of WebRTC to the unique challenges of international deployments, codec differences, and the open-source revolution in telephony. Learn how traditional telephony infrastructure is being replaced by scalable, IP-based systems and what the future holds for the industry. Whether you're an industry veteran or a curious newbie, this episode offers valuable insights into the telephony landscape.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Nikola welcomes Oisin Glynn, VP of Solution Consulting, to discuss the fascinating history and evolution of modern telephony. Going back to the inception of telephone systems, from their early manual connections to the adoption of the PSTN and the advent of private branch exchanges (PBX), the conversation touches on the impact of the internet and mobile telephony on traditional phone systems, the rise of VoIP, and how companies like Cisco and Avaya played pivotal roles in transforming business communications. Whether you're curious about the historical milestones or the technical nuances of modern telecommunication, this episode is a perfect start to a series on the wizardry in present day telephonics.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

** Check out the report that inspired the episode **

Damien welcomes PolyAI's senior director of brand, Kylie Whitehead, to discuss the findings from two comprehensive surveys highlighting trends in customer experience (CX) and contact center leadership. Learn about the evolving roles of CX leaders and the increasing influence of AI, as well as the differing sentiments of contact center agents towards automation.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Bonus breaking news episode of the Deep Learning with PolyAI podcast! Hosts Damien and Nikola chat about the recent release of DeepSeek's open-source R1 model, which has sparked widespread reaction across the tech world. The R1 model, created with limited resources, has matched or exceeded benchmarks of leading models from OpenAI and Meta, prompting discussions on its potential impact on AI development and market dynamics. The episode explores DeepSeek's engineering feats, the subsequent market responses, and its implications for the future of open-source AI movements and global competition.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Kylie takes the helm for this episode of Deep Learning with Poly AI, where she and PolyAI CEO Nikola Mrkšić chat(no GPT) about OpenAI's latest innovation: the Operator agent. From the concept of AI performing tasks such as ordering groceries, booking travel, and making restaurant reservations, to live demos of agentic requests, Kylie and Nikola butt heads on overall impact and explore the broader implications of AI in increasing productivity, interface interaction, and transforming enterprise applications. 

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube

Damien and Nikola (CEO & Co-Founder of PolyAI) talk about the recently announced AI Opportunities Action Plan. The conversation highlights the UK’s historical contributions to AI, the challenges it faces due to American tech dominance, and the strategic moves to build a competitive AI ecosystem. They touch on the importance of infrastructure, talent attraction, public sector engagement, and ethical considerations in AI development. Tune in to hear detailed insights on how this plan could impact the startup scene, established industries, and the broader economic landscape.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube