Certification and Training

Genesys Education is here, along with some experts from the Genesys Product Team to share insights on LAM powered Agentic Virtual Agents, a new approach to virtual self-service that can reason, take action, and move customers toward outcomes, not just respond to questions.  

If you're evaluating AI investments, wondering if this is right for your business, or thinking about how to get started, read on.  

Q1: What's the #1 thing new users should understand about LAM Agentic Virtual Agents? 

LAM Agentic Virtual Agents aren't just more conversational bots, they're designed to work toward an outcome. Instead of following a fixed flow or just answering questions, they can: 

• Understand a customer's goal
• Decide what steps are needed 
• Use tools and knowledge to take action 

That shift, from responding to reasoning + acting, is what makes them different. 

Q2: When are LAM Agentic Virtual Agents the right fit? 

LAM Agentic Virtual Agents are a strong fit when a customer interaction requires both conversation and action, especially when the path isn't predictable. 

Good examples: 

• Checking eligibility, then applying a change 
• Pulling info from multiple systems, then acting on it 
• Handling requests that evolve mid-conversation 

Not every use case needs this level of autonomy. 

• If you mainly need Q&A from knowledge, an LLM-powered virtual agent may be enough 
• If you need strict, repeatable workflows, AI Guides may be a better fit 

The key question: Do you need the system to decide and act, or just respond? 

Q3: What should decision makers think about before investing? 

Start with the problem, not the technology. 

LAM Agentic Virtual Agents work best when: 

• The customer goal is valuable to automate 
• The journey has enough variability that rigid flows break down 
• You can clearly define what the agent should and shouldn't do 

Also think about: 

• What systems or data it needs access to 
• What actions should stay human-led 
• How you'll measure success (containment, speed, effort, etc.) 

A focused first use case usually beats a broad one. 

Q4: What do admins and architects need to plan for? 

The experience may feel simple, but good setup still matters. 

In Genesys Cloud, you'll configure: 

• The agent's role and guidelines that keep it aligned to policy 
• The tools it can use (actions across systems) 
• The knowledge it can draw from 
• The guardrails that prevent unwanted behavior 

Then you connect it into Architect using a Virtual Agent flow where you can also determine additional aspects such as what information can be shared 

So, the work shifts from building hundreds of steps to making smart decisions about scope, access, and boundaries. 

Q5: What are some key considerations when starting with LAM Powered Agentic Virtual Agents?

Early decisions and focus areas that can shape your success:

• Start with a focused, high-value journey rather than casting too wide a net early on
• Invest in the right tools and knowledge to fully enable what the agent can accomplish
• Design with clear outcomes in mind, not just the conversation itself
• Evaluate the fit upfront to ensure LAM Agentic Virtual Agents are the right approach versus a simpler solution

Q6: Do you have a favorite tip for getting started? 

Yes: design around a customer outcome, not the feature set. Pick one journey where: 

• Customers struggle today 
• Multiple steps or systems are involved 
• Better reasoning + action would clearly help 

Start there, test, refine, and then expand. 

Q7: Where can learners go for more info? 

• Start with the LAM Agentic Virtual Agent Introduction eLearning for the big picture 
• Explore the Resource Center documentation for setup and capabilities 
• Watch the configuration webinar for a practical walkthrough 

Join the discussion 

What are you trying to solve with AI in your organization? 

• Are you evaluating whether LAM Agentic Virtual Agent is the right fit? 
• Do you have a use case in mind but aren't sure where it fits? 
• Curious how others are getting started? 

Drop your questions below, our experts are here to help. 

Genesys Education is here, along with some experts from the Genesys Product Team to share insights on LAM powered Agentic Virtual Agents, a new approach to virtual self-service that can reason, take action, and move customers toward outcomes, not just respond to questions.  

If you're evaluating AI investments, wondering if this is right for your business, or thinking about how to get started, read on.  

Q1: What's the #1 thing new users should understand about LAM Agentic Virtual Agents? 

LAM Agentic Virtual Agents aren't just more conversational bots, they're designed to work toward an outcome. Instead of following a fixed flow or just answering questions, they can: 

• Understand a customer's goal
• Decide what steps are needed 
• Use tools and knowledge to take action 

That shift, from responding to reasoning + acting, is what makes them different. 

Q2: When are LAM Agentic Virtual Agents the right fit? 

LAM Agentic Virtual Agents are a strong fit when a customer interaction requires both conversation and action, especially when the path isn't predictable. 

Good examples: 

• Checking eligibility, then applying a change 
• Pulling info from multiple systems, then acting on it 
• Handling requests that evolve mid-conversation 

Not every use case needs this level of autonomy. 

• If you mainly need Q&A from knowledge, an LLM-powered virtual agent may be enough 
• If you need strict, repeatable workflows, AI Guides may be a better fit 

The key question: Do you need the system to decide and act, or just respond? 

Q3: What should decision makers think about before investing? 

Start with the problem, not the technology. 

LAM Agentic Virtual Agents work best when: 

• The customer goal is valuable to automate 
• The journey has enough variability that rigid flows break down 
• You can clearly define what the agent should and shouldn't do 

Also think about: 

• What systems or data it needs access to 
• What actions should stay human-led 
• How you'll measure success (containment, speed, effort, etc.) 

A focused first use case usually beats a broad one. 

Q4: What do admins and architects need to plan for? 

The experience may feel simple, but good setup still matters. 

In Genesys Cloud, you'll configure: 

• The agent's role and guidelines that keep it aligned to policy 
• The tools it can use (actions across systems) 
• The knowledge it can draw from 
• The guardrails that prevent unwanted behavior 

Then you connect it into Architect using a Virtual Agent flow where you can also determine additional aspects such as what information can be shared 

So, the work shifts from building hundreds of steps to making smart decisions about scope, access, and boundaries. 

Q5: What are some key considerations when starting with LAM Powered Agentic Virtual Agents?

Early decisions and focus areas that can shape your success:

• Start with a focused, high-value journey rather than casting too wide a net early on
• Invest in the right tools and knowledge to fully enable what the agent can accomplish
• Design with clear outcomes in mind, not just the conversation itself
• Evaluate the fit upfront to ensure LAM Agentic Virtual Agents are the right approach versus a simpler solution

Q6: Do you have a favorite tip for getting started? 

Yes: design around a customer outcome, not the feature set. Pick one journey where: 

• Customers struggle today 
• Multiple steps or systems are involved 
• Better reasoning + action would clearly help 

Start there, test, refine, and then expand. 

Q7: Where can learners go for more info? 

• Start with the LAM Agentic Virtual Agent Introduction eLearning for the big picture 
• Explore the Resource Center documentation for setup and capabilities 
• Watch the configuration webinar for a practical walkthrough 

Join the discussion 

What are you trying to solve with AI in your organization? 

• Are you evaluating whether LAM Agentic Virtual Agent is the right fit? 
• Do you have a use case in mind but aren't sure where it fits? 
• Curious how others are getting started? 

Drop your questions below, our experts are here to help. 

Genesys Education is here, along with some experts from the Genesys Product Team to share insights on LAM powered Agentic Virtual Agents, a new approach to virtual self-service that can reason, take action, and move customers toward outcomes, not just respond to questions.  

If you're evaluating AI investments, wondering if this is right for your business, or thinking about how to get started, read on.  

Q1: What's the #1 thing new users should understand about LAM Agentic Virtual Agents? 

LAM Agentic Virtual Agents aren't just more conversational bots, they're designed to work toward an outcome. Instead of following a fixed flow or just answering questions, they can: 

• Understand a customer's goal
• Decide what steps are needed 
• Use tools and knowledge to take action 

That shift, from responding to reasoning + acting, is what makes them different. 

Q2: When are LAM Agentic Virtual Agents the right fit? 

LAM Agentic Virtual Agents are a strong fit when a customer interaction requires both conversation and action, especially when the path isn't predictable. 

Good examples: 

• Checking eligibility, then applying a change 
• Pulling info from multiple systems, then acting on it 
• Handling requests that evolve mid-conversation 

Not every use case needs this level of autonomy. 

• If you mainly need Q&A from knowledge, an LLM-powered virtual agent may be enough 
• If you need strict, repeatable workflows, AI Guides may be a better fit 

The key question: Do you need the system to decide and act, or just respond? 

Q3: What should decision makers think about before investing? 

Start with the problem, not the technology. 

LAM Agentic Virtual Agents work best when: 

• The customer goal is valuable to automate 
• The journey has enough variability that rigid flows break down 
• You can clearly define what the agent should and shouldn't do 

Also think about: 

• What systems or data it needs access to 
• What actions should stay human-led 
• How you'll measure success (containment, speed, effort, etc.) 

A focused first use case usually beats a broad one. 

Q4: What do admins and architects need to plan for? 

The experience may feel simple, but good setup still matters. 

In Genesys Cloud, you'll configure: 

• The agent's role and guidelines that keep it aligned to policy 
• The tools it can use (actions across systems) 
• The knowledge it can draw from 
• The guardrails that prevent unwanted behavior 

Then you connect it into Architect using a Virtual Agent flow where you can also determine additional aspects such as what information can be shared 

So, the work shifts from building hundreds of steps to making smart decisions about scope, access, and boundaries. 

Q5: What are some key considerations when starting with LAM Powered Agentic Virtual Agents?

Early decisions and focus areas that can shape your success:

• Start with a focused, high-value journey rather than casting too wide a net early on
• Invest in the right tools and knowledge to fully enable what the agent can accomplish
• Design with clear outcomes in mind, not just the conversation itself
• Evaluate the fit upfront to ensure LAM Agentic Virtual Agents are the right approach versus a simpler solution

Q6: Do you have a favorite tip for getting started? 

Yes: design around a customer outcome, not the feature set. Pick one journey where: 

• Customers struggle today 
• Multiple steps or systems are involved 
• Better reasoning + action would clearly help 

Start there, test, refine, and then expand. 

Q7: Where can learners go for more info? 

• Start with the LAM Agentic Virtual Agent Introduction eLearning for the big picture 
• Explore the Resource Center documentation for setup and capabilities 
• Watch the configuration webinar for a practical walkthrough 

Join the discussion 

What are you trying to solve with AI in your organization? 

• Are you evaluating whether LAM Agentic Virtual Agent is the right fit? 
• Do you have a use case in mind but aren't sure where it fits? 
• Curious how others are getting started? 

Drop your questions below, our experts are here to help. 

Genesys Education is here, along with some experts from the Genesys Product Team to share insights on LAM powered Agentic Virtual Agents, a new approach to virtual self-service that can reason, take action, and move customers toward outcomes, not just respond to questions.  

If you're evaluating AI investments, wondering if this is right for your business, or thinking about how to get started, read on.  

Q1: What's the #1 thing new users should understand about LAM Agentic Virtual Agents? 

LAM Agentic Virtual Agents aren't just more conversational bots, they're designed to work toward an outcome. Instead of following a fixed flow or just answering questions, they can: 

• Understand a customer's goal
• Decide what steps are needed 
• Use tools and knowledge to take action 

That shift, from responding to reasoning + acting, is what makes them different. 

Q2: When are LAM Agentic Virtual Agents the right fit? 

LAM Agentic Virtual Agents are a strong fit when a customer interaction requires both conversation and action, especially when the path isn't predictable. 

Good examples: 

• Checking eligibility, then applying a change 
• Pulling info from multiple systems, then acting on it 
• Handling requests that evolve mid-conversation 

Not every use case needs this level of autonomy. 

• If you mainly need Q&A from knowledge, an LLM-powered virtual agent may be enough 
• If you need strict, repeatable workflows, AI Guides may be a better fit 

The key question: Do you need the system to decide and act, or just respond? 

Q3: What should decision makers think about before investing? 

Start with the problem, not the technology. 

LAM Agentic Virtual Agents work best when: 

• The customer goal is valuable to automate 
• The journey has enough variability that rigid flows break down 
• You can clearly define what the agent should and shouldn't do 

Also think about: 

• What systems or data it needs access to 
• What actions should stay human-led 
• How you'll measure success (containment, speed, effort, etc.) 

A focused first use case usually beats a broad one. 

Q4: What do admins and architects need to plan for? 

The experience may feel simple, but good setup still matters. 

In Genesys Cloud, you'll configure: 

• The agent's role and guidelines that keep it aligned to policy 
• The tools it can use (actions across systems) 
• The knowledge it can draw from 
• The guardrails that prevent unwanted behavior 

Then you connect it into Architect using a Virtual Agent flow where you can also determine additional aspects such as what information can be shared 

So, the work shifts from building hundreds of steps to making smart decisions about scope, access, and boundaries. 

Q5: What are some key considerations when starting with LAM Powered Agentic Virtual Agents?

Early decisions and focus areas that can shape your success:

• Start with a focused, high-value journey rather than casting too wide a net early on
• Invest in the right tools and knowledge to fully enable what the agent can accomplish
• Design with clear outcomes in mind, not just the conversation itself
• Evaluate the fit upfront to ensure LAM Agentic Virtual Agents are the right approach versus a simpler solution

Q6: Do you have a favorite tip for getting started? 

Yes: design around a customer outcome, not the feature set. Pick one journey where: 

• Customers struggle today 
• Multiple steps or systems are involved 
• Better reasoning + action would clearly help 

Start there, test, refine, and then expand. 

Q7: Where can learners go for more info? 

• Start with the LAM Agentic Virtual Agent Introduction eLearning for the big picture 
• Explore the Resource Center documentation for setup and capabilities 
• Watch the configuration webinar for a practical walkthrough 

Join the discussion 

What are you trying to solve with AI in your organization? 

• Are you evaluating whether LAM Agentic Virtual Agent is the right fit? 
• Do you have a use case in mind but aren't sure where it fits? 
• Curious how others are getting started? 

Drop your questions below, our experts are here to help. 

Genesys Education is here, along with some experts from the Genesys Product Team to share insights on LAM powered Agentic Virtual Agents, a new approach to virtual self-service that can reason, take action, and move customers toward outcomes, not just respond to questions.  

If you're evaluating AI investments, wondering if this is right for your business, or thinking about how to get started, read on.  

Q1: What's the #1 thing new users should understand about LAM Agentic Virtual Agents? 

LAM Agentic Virtual Agents aren't just more conversational bots, they're designed to work toward an outcome. Instead of following a fixed flow or just answering questions, they can: 

• Understand a customer's goal
• Decide what steps are needed 
• Use tools and knowledge to take action 

That shift, from responding to reasoning + acting, is what makes them different. 

Q2: When are LAM Agentic Virtual Agents the right fit? 

LAM Agentic Virtual Agents are a strong fit when a customer interaction requires both conversation and action, especially when the path isn't predictable. 

Good examples: 

• Checking eligibility, then applying a change 
• Pulling info from multiple systems, then acting on it 
• Handling requests that evolve mid-conversation 

Not every use case needs this level of autonomy. 

• If you mainly need Q&A from knowledge, an LLM-powered virtual agent may be enough 
• If you need strict, repeatable workflows, AI Guides may be a better fit 

The key question: Do you need the system to decide and act, or just respond? 

Q3: What should decision makers think about before investing? 

Start with the problem, not the technology. 

LAM Agentic Virtual Agents work best when: 

• The customer goal is valuable to automate 
• The journey has enough variability that rigid flows break down 
• You can clearly define what the agent should and shouldn't do 

Also think about: 

• What systems or data it needs access to 
• What actions should stay human-led 
• How you'll measure success (containment, speed, effort, etc.) 

A focused first use case usually beats a broad one. 

Q4: What do admins and architects need to plan for? 

The experience may feel simple, but good setup still matters. 

In Genesys Cloud, you'll configure: 

• The agent's role and guidelines that keep it aligned to policy 
• The tools it can use (actions across systems) 
• The knowledge it can draw from 
• The guardrails that prevent unwanted behavior 

Then you connect it into Architect using a Virtual Agent flow where you can also determine additional aspects such as what information can be shared 

So, the work shifts from building hundreds of steps to making smart decisions about scope, access, and boundaries. 

Q5: What are some key considerations when starting with LAM Powered Agentic Virtual Agents?

Early decisions and focus areas that can shape your success:

• Start with a focused, high-value journey rather than casting too wide a net early on
• Invest in the right tools and knowledge to fully enable what the agent can accomplish
• Design with clear outcomes in mind, not just the conversation itself
• Evaluate the fit upfront to ensure LAM Agentic Virtual Agents are the right approach versus a simpler solution

Q6: Do you have a favorite tip for getting started? 

Yes: design around a customer outcome, not the feature set. Pick one journey where: 

• Customers struggle today 
• Multiple steps or systems are involved 
• Better reasoning + action would clearly help 

Start there, test, refine, and then expand. 

Q7: Where can learners go for more info? 

• Start with the LAM Agentic Virtual Agent Introduction eLearning for the big picture 
• Explore the Resource Center documentation for setup and capabilities 
• Watch the configuration webinar for a practical walkthrough 

Join the discussion 

What are you trying to solve with AI in your organization? 

• Are you evaluating whether LAM Agentic Virtual Agent is the right fit? 
• Do you have a use case in mind but aren't sure where it fits? 
• Curious how others are getting started? 

Drop your questions below, our experts are here to help. 

Genesys Education is here, along with some experts from the Genesys Product Team to share insights on LAM powered Agentic Virtual Agents, a new approach to virtual self-service that can reason, take action, and move customers toward outcomes, not just respond to questions.  

If you're evaluating AI investments, wondering if this is right for your business, or thinking about how to get started, read on.  

Q1: What's the #1 thing new users should understand about LAM Agentic Virtual Agents? 

LAM Agentic Virtual Agents aren't just more conversational bots, they're designed to work toward an outcome. Instead of following a fixed flow or just answering questions, they can: 

• Understand a customer's goal
• Decide what steps are needed 
• Use tools and knowledge to take action 

That shift, from responding to reasoning + acting, is what makes them different. 

Q2: When are LAM Agentic Virtual Agents the right fit? 

LAM Agentic Virtual Agents are a strong fit when a customer interaction requires both conversation and action, especially when the path isn't predictable. 

Good examples: 

• Checking eligibility, then applying a change 
• Pulling info from multiple systems, then acting on it 
• Handling requests that evolve mid-conversation 

Not every use case needs this level of autonomy. 

• If you mainly need Q&A from knowledge, an LLM-powered virtual agent may be enough 
• If you need strict, repeatable workflows, AI Guides may be a better fit 

The key question: Do you need the system to decide and act, or just respond? 

Q3: What should decision makers think about before investing? 

Start with the problem, not the technology. 

LAM Agentic Virtual Agents work best when: 

• The customer goal is valuable to automate 
• The journey has enough variability that rigid flows break down 
• You can clearly define what the agent should and shouldn't do 

Also think about: 

• What systems or data it needs access to 
• What actions should stay human-led 
• How you'll measure success (containment, speed, effort, etc.) 

A focused first use case usually beats a broad one. 

Q4: What do admins and architects need to plan for? 

The experience may feel simple, but good setup still matters. 

In Genesys Cloud, you'll configure: 

• The agent's role and guidelines that keep it aligned to policy 
• The tools it can use (actions across systems) 
• The knowledge it can draw from 
• The guardrails that prevent unwanted behavior 

Then you connect it into Architect using a Virtual Agent flow where you can also determine additional aspects such as what information can be shared 

So, the work shifts from building hundreds of steps to making smart decisions about scope, access, and boundaries. 

Q5: What are some key considerations when starting with LAM Powered Agentic Virtual Agents?

Early decisions and focus areas that can shape your success:

• Start with a focused, high-value journey rather than casting too wide a net early on
• Invest in the right tools and knowledge to fully enable what the agent can accomplish
• Design with clear outcomes in mind, not just the conversation itself
• Evaluate the fit upfront to ensure LAM Agentic Virtual Agents are the right approach versus a simpler solution

Q6: Do you have a favorite tip for getting started? 

Yes: design around a customer outcome, not the feature set. Pick one journey where: 

• Customers struggle today 
• Multiple steps or systems are involved 
• Better reasoning + action would clearly help 

Start there, test, refine, and then expand. 

Q7: Where can learners go for more info? 

• Start with the LAM Agentic Virtual Agent Introduction eLearning for the big picture 
• Explore the Resource Center documentation for setup and capabilities 
• Watch the configuration webinar for a practical walkthrough 

Join the discussion 

What are you trying to solve with AI in your organization? 

• Are you evaluating whether LAM Agentic Virtual Agent is the right fit? 
• Do you have a use case in mind but aren't sure where it fits? 
• Curious how others are getting started? 

Drop your questions below, our experts are here to help. 

Genesys Education is here, along with some experts from the Genesys Product Team to share insights on LAM powered Agentic Virtual Agents, a new approach to virtual self-service that can reason, take action, and move customers toward outcomes, not just respond to questions.  

If you're evaluating AI investments, wondering if this is right for your business, or thinking about how to get started, read on.  

Q1: What's the #1 thing new users should understand about LAM Agentic Virtual Agents? 

LAM Agentic Virtual Agents aren't just more conversational bots, they're designed to work toward an outcome. Instead of following a fixed flow or just answering questions, they can: 

• Understand a customer's goal
• Decide what steps are needed 
• Use tools and knowledge to take action 

That shift, from responding to reasoning + acting, is what makes them different. 

Q2: When are LAM Agentic Virtual Agents the right fit? 

LAM Agentic Virtual Agents are a strong fit when a customer interaction requires both conversation and action, especially when the path isn't predictable. 

Good examples: 

• Checking eligibility, then applying a change 
• Pulling info from multiple systems, then acting on it 
• Handling requests that evolve mid-conversation 

Not every use case needs this level of autonomy. 

• If you mainly need Q&A from knowledge, an LLM-powered virtual agent may be enough 
• If you need strict, repeatable workflows, AI Guides may be a better fit 

The key question: Do you need the system to decide and act, or just respond? 

Q3: What should decision makers think about before investing? 

Start with the problem, not the technology. 

LAM Agentic Virtual Agents work best when: 

• The customer goal is valuable to automate 
• The journey has enough variability that rigid flows break down 
• You can clearly define what the agent should and shouldn't do 

Also think about: 

• What systems or data it needs access to 
• What actions should stay human-led 
• How you'll measure success (containment, speed, effort, etc.) 

A focused first use case usually beats a broad one. 

Q4: What do admins and architects need to plan for? 

The experience may feel simple, but good setup still matters. 

In Genesys Cloud, you'll configure: 

• The agent's role and guidelines that keep it aligned to policy 
• The tools it can use (actions across systems) 
• The knowledge it can draw from 
• The guardrails that prevent unwanted behavior 

Then you connect it into Architect using a Virtual Agent flow where you can also determine additional aspects such as what information can be shared 

So, the work shifts from building hundreds of steps to making smart decisions about scope, access, and boundaries. 

Q5: What are some key considerations when starting with LAM Powered Agentic Virtual Agents?

Early decisions and focus areas that can shape your success:

• Start with a focused, high-value journey rather than casting too wide a net early on
• Invest in the right tools and knowledge to fully enable what the agent can accomplish
• Design with clear outcomes in mind, not just the conversation itself
• Evaluate the fit upfront to ensure LAM Agentic Virtual Agents are the right approach versus a simpler solution

Q6: Do you have a favorite tip for getting started? 

Yes: design around a customer outcome, not the feature set. Pick one journey where: 

• Customers struggle today 
• Multiple steps or systems are involved 
• Better reasoning + action would clearly help 

Start there, test, refine, and then expand. 

Q7: Where can learners go for more info? 

• Start with the LAM Agentic Virtual Agent Introduction eLearning for the big picture 
• Explore the Resource Center documentation for setup and capabilities 
• Watch the configuration webinar for a practical walkthrough 

Join the discussion 

What are you trying to solve with AI in your organization? 

• Are you evaluating whether LAM Agentic Virtual Agent is the right fit? 
• Do you have a use case in mind but aren't sure where it fits? 
• Curious how others are getting started? 

Drop your questions below, our experts are here to help. 

Genesys Education is here, along with some experts from the Genesys Product Team to share insights on LAM powered Agentic Virtual Agents, a new approach to virtual self-service that can reason, take action, and move customers toward outcomes, not just respond to questions.  

Expert: Mitchell Mason - Principal Product Manager, Virtual Agent

If you're evaluating AI investments, wondering if this is right for your business, or thinking about how to get started, read on.  

Q1: What's the #1 thing new users should understand about LAM Agentic Virtual Agents? 

LAM Agentic Virtual Agents aren't just more conversational bots, they're designed to work toward an outcome. Instead of following a fixed flow or just answering questions, they can: 

• Understand a customer's goal
• Decide what steps are needed 
• Use tools and knowledge to take action 

That shift, from responding to reasoning + acting, is what makes them different. 

Q2: When are LAM Agentic Virtual Agents the right fit? 

LAM Agentic Virtual Agents are a strong fit when a customer interaction requires both conversation and action, especially when the path isn't predictable. 

Good examples: 

• Checking eligibility, then applying a change 
• Pulling info from multiple systems, then acting on it 
• Handling requests that evolve mid-conversation 

Not every use case needs this level of autonomy. 

• If you mainly need Q&A from knowledge, an LLM-powered virtual agent may be enough 
• If you need strict, repeatable workflows, AI Guides may be a better fit 

The key question: Do you need the system to decide and act, or just respond? 

Q3: What should decision makers think about before investing? 

Start with the problem, not the technology. 

LAM Agentic Virtual Agents work best when: 

• The customer goal is valuable to automate 
• The journey has enough variability that rigid flows break down 
• You can clearly define what the agent should and shouldn't do 

Also think about: 

• What systems or data it needs access to 
• What actions should stay human-led 
• How you'll measure success (containment, speed, effort, etc.) 

A focused first use case usually beats a broad one. 

Q4: What do admins and architects need to plan for? 

The experience may feel simple, but good setup still matters. 

In Genesys Cloud, you'll configure: 

• The agent's role and guidelines that keep it aligned to policy 
• The tools it can use (actions across systems) 
• The knowledge it can draw from 
• The guardrails that prevent unwanted behavior 

Then you connect it into Architect using a Virtual Agent flow where you can also determine additional aspects such as what information can be shared 

So, the work shifts from building hundreds of steps to making smart decisions about scope, access, and boundaries. 

Q5: What are some key considerations when starting with LAM Powered Agentic Virtual Agents?

Early decisions and focus areas that can shape your success:

• Start with a focused, high-value journey rather than casting too wide a net early on
• Invest in the right tools and knowledge to fully enable what the agent can accomplish
• Design with clear outcomes in mind, not just the conversation itself
• Evaluate the fit upfront to ensure LAM Agentic Virtual Agents are the right approach versus a simpler solution

Q6: Do you have a favorite tip for getting started? 

Yes: design around a customer outcome, not the feature set. Pick one journey where: 

• Customers struggle today 
• Multiple steps or systems are involved 
• Better reasoning + action would clearly help 

Start there, test, refine, and then expand. 

Q7: Where can learners go for more info? 

• Start with the LAM Agentic Virtual Agent Introduction eLearning for the big picture 
• Explore the Resource Center documentation for setup and capabilities 
• Watch the configuration webinar for a practical walkthrough 

Join the discussion 

What are you trying to solve with AI in your organization? 

• Are you evaluating whether LAM Agentic Virtual Agent is the right fit? 
• Do you have a use case in mind but aren't sure where it fits? 
• Curious how others are getting started? 

Drop your questions below, our experts are here to help. 

Genesys Education is here, along with some experts from the Genesys Product Team to share insights on LAM powered Agentic Virtual Agents, a new approach to virtual self-service that can reason, take action, and move customers toward outcomes, not just respond to questions.  

Expert: Mitchell Mason - Principal Product Manager, Virtual Agent

If you're evaluating AI investments, wondering if this is right for your business, or thinking about how to get started, read on.  

Q1: What's the #1 thing new users should understand about LAM Agentic Virtual Agents? 

LAM Agentic Virtual Agents aren't just more conversational bots, they're designed to work toward an outcome. Instead of following a fixed flow or just answering questions, they can: 

• Understand a customer's goal
• Decide what steps are needed 
• Use tools and knowledge to take action 

That shift, from responding to reasoning + acting, is what makes them different. 

Q2: When are LAM Agentic Virtual Agents the right fit? 

LAM Agentic Virtual Agents are a strong fit when a customer interaction requires both conversation and action, especially when the path isn't predictable. 

Good examples: 

• Checking eligibility, then applying a change 
• Pulling info from multiple systems, then acting on it 
• Handling requests that evolve mid-conversation 

Not every use case needs this level of autonomy. 

• If you mainly need Q&A from knowledge, an LLM-powered virtual agent may be enough 
• If you need strict, repeatable workflows, AI Guides may be a better fit 

The key question: Do you need the system to decide and act, or just respond? 

Q3: What should decision makers think about before investing? 

Start with the problem, not the technology. 

LAM Agentic Virtual Agents work best when: 

• The customer goal is valuable to automate 
• The journey has enough variability that rigid flows break down 
• You can clearly define what the agent should and shouldn't do 

Also think about: 

• What systems or data it needs access to 
• What actions should stay human-led 
• How you'll measure success (containment, speed, effort, etc.) 

A focused first use case usually beats a broad one. 

Q4: What do admins and architects need to plan for? 

The experience may feel simple, but good setup still matters. 

In Genesys Cloud, you'll configure: 

• The agent's role and guidelines that keep it aligned to policy 
• The tools it can use (actions across systems) 
• The knowledge it can draw from 
• The guardrails that prevent unwanted behavior 

Then you connect it into Architect using a Virtual Agent flow where you can also determine additional aspects such as what information can be shared 

So, the work shifts from building hundreds of steps to making smart decisions about scope, access, and boundaries. 

Q5: What are some key considerations when starting with LAM Powered Agentic Virtual Agents?

Early decisions and focus areas that can shape your success:

• Start with a focused, high-value journey rather than casting too wide a net early on
• Invest in the right tools and knowledge to fully enable what the agent can accomplish
• Design with clear outcomes in mind, not just the conversation itself
• Evaluate the fit upfront to ensure LAM Agentic Virtual Agents are the right approach versus a simpler solution

Q6: Do you have a favorite tip for getting started? 

Yes: design around a customer outcome, not the feature set. Pick one journey where: 

• Customers struggle today 
• Multiple steps or systems are involved 
• Better reasoning + action would clearly help 

Start there, test, refine, and then expand. 

Q7: Where can learners go for more info? 

• Start with the LAM Agentic Virtual Agent Introduction eLearning for the big picture 
• Explore the Resource Center documentation for setup and capabilities 
• Watch the configuration webinar for a practical walkthrough 

Join the discussion 

What are you trying to solve with AI in your organization? 

• Are you evaluating whether LAM Agentic Virtual Agent is the right fit? 
• Do you have a use case in mind but aren't sure where it fits? 
• Curious how others are getting started? 

Drop your questions below, our experts are here to help. 

Genesys Education is here, along with some experts from the Genesys Product Team to share insights on LAM powered Agentic Virtual Agents, a new approach to virtual self-service that can reason, take action, and move customers toward outcomes, not just respond to questions.  

Expert: Mitchell Mason - Principal Product Manager, Virtual Agent

If you're evaluating AI investments, wondering if this is right for your business, or thinking about how to get started, read on.  

Q1: What's the #1 thing new users should understand about LAM Agentic Virtual Agents? 

LAM Agentic Virtual Agents aren't just more conversational bots, they're designed to work toward an outcome. Instead of following a fixed flow or just answering questions, they can: 

• Understand a customer's goal
• Decide what steps are needed 
• Use tools and knowledge to take action 

That shift, from responding to reasoning + acting, is what makes them different. 

Q2: When are LAM Agentic Virtual Agents the right fit? 

LAM Agentic Virtual Agents are a strong fit when a customer interaction requires both conversation and action, especially when the path isn't predictable. 

Good examples: 

• Checking eligibility, then applying a change 
• Pulling info from multiple systems, then acting on it 
• Handling requests that evolve mid-conversation 

Not every use case needs this level of autonomy. 

• If you mainly need Q&A from knowledge, an LLM-powered virtual agent may be enough 
• If you need strict, repeatable workflows, AI Guides may be a better fit 

The key question: Do you need the system to decide and act, or just respond? 

Q3: What should decision makers think about before investing? 

Start with the problem, not the technology. 

LAM Agentic Virtual Agents work best when: 

• The customer goal is valuable to automate 
• The journey has enough variability that rigid flows break down 
• You can clearly define what the agent should and shouldn't do 

Also think about: 

• What systems or data it needs access to 
• What actions should stay human-led 
• How you'll measure success (containment, speed, effort, etc.) 

A focused first use case usually beats a broad one. 

Q4: What do admins and architects need to plan for? 

The experience may feel simple, but good setup still matters. 

In Genesys Cloud, you'll configure: 

• The agent's role and guidelines that keep it aligned to policy 
• The tools it can use (actions across systems) 
• The knowledge it can draw from 
• The guardrails that prevent unwanted behavior 

Then you connect it into Architect using a Virtual Agent flow where you can also determine additional aspects such as what information can be shared 

So, the work shifts from building hundreds of steps to making smart decisions about scope, access, and boundaries. 

Q5: What are some key considerations when starting with LAM Powered Agentic Virtual Agents?

Early decisions and focus areas that can shape your success:

• Start with a focused, high-value journey rather than casting too wide a net early on
• Invest in the right tools and knowledge to fully enable what the agent can accomplish
• Design with clear outcomes in mind, not just the conversation itself
• Evaluate the fit upfront to ensure LAM Agentic Virtual Agents are the right approach versus a simpler solution

Q6: Do you have a favorite tip for getting started? 

Yes: design around a customer outcome, not the feature set. Pick one journey where: 

• Customers struggle today 
• Multiple steps or systems are involved 
• Better reasoning + action would clearly help 

Start there, test, refine, and then expand. 

Q7: Where can learners go for more info? 

• Start with the LAM Agentic Virtual Agent Introduction eLearning for the big picture 
• Explore the Resource Center documentation for setup and capabilities 
• Watch the configuration webinar for a practical walkthrough 

Join the discussion 

What are you trying to solve with AI in your organization? 

• Are you evaluating whether LAM Agentic Virtual Agent is the right fit? 
• Do you have a use case in mind but aren't sure where it fits? 
• Curious how others are getting started? 

Drop your questions below, our experts are here to help. 

Genesys Education is here, along with some experts from the Genesys Product Team to share insights on LAM powered Agentic Virtual Agents, a new approach to virtual self-service that can reason, take action, and move customers toward outcomes, not just respond to questions.  

Expert: Mitchell Mason - Principal Product Manager, Virtual Agent

If you're evaluating AI investments, wondering if this is right for your business, or thinking about how to get started, read on.  

Q1: What's the #1 thing new users should understand about LAM Agentic Virtual Agents? 

LAM Agentic Virtual Agents aren't just more conversational bots, they're designed to work toward an outcome. Instead of following a fixed flow or just answering questions, they can: 

• Understand a customer's goal
• Decide what steps are needed 
• Use tools and knowledge to take action 

That shift, from responding to reasoning + acting, is what makes them different. 

Q2: When are LAM Agentic Virtual Agents the right fit? 

LAM Agentic Virtual Agents are a strong fit when a customer interaction requires both conversation and action, especially when the path isn't predictable. 

Good examples: 

• Checking eligibility, then applying a change 
• Pulling info from multiple systems, then acting on it 
• Handling requests that evolve mid-conversation 

Not every use case needs this level of autonomy. 

• If you mainly need Q&A from knowledge, an LLM-powered virtual agent may be enough 
• If you need strict, repeatable workflows, AI Guides may be a better fit 

The key question: Do you need the system to decide and act, or just respond? 

Q3: What should decision makers think about before investing? 

Start with the problem, not the technology. 

LAM Agentic Virtual Agents work best when: 

• The customer goal is valuable to automate 
• The journey has enough variability that rigid flows break down 
• You can clearly define what the agent should and shouldn't do 

Also think about: 

• What systems or data it needs access to 
• What actions should stay human-led 
• How you'll measure success (containment, speed, effort, etc.) 

A focused first use case usually beats a broad one. 

Q4: What do admins and architects need to plan for? 

The experience may feel simple, but good setup still matters. 

In Genesys Cloud, you'll configure: 

• The agent's role and guidelines that keep it aligned to policy 
• The tools it can use (actions across systems) 
• The knowledge it can draw from 
• The guardrails that prevent unwanted behavior 

Then you connect it into Architect using a Virtual Agent flow where you can also determine additional aspects such as what information can be shared 

So, the work shifts from building hundreds of steps to making smart decisions about scope, access, and boundaries. 

Q5: What are some key considerations when starting with LAM Powered Agentic Virtual Agents?

Early decisions and focus areas that can shape your success:

• Start with a focused, high-value journey rather than casting too wide a net early on
• Invest in the right tools and knowledge to fully enable what the agent can accomplish
• Design with clear outcomes in mind, not just the conversation itself
• Evaluate the fit upfront to ensure LAM Agentic Virtual Agents are the right approach versus a simpler solution

Q6: Do you have a favorite tip for getting started? 

Yes: design around a customer outcome, not the feature set. Pick one journey where: 

• Customers struggle today 
• Multiple steps or systems are involved 
• Better reasoning + action would clearly help 

Start there, test, refine, and then expand. 

Q7: Where can learners go for more info? 

• Start with the LAM Agentic Virtual Agent Introduction eLearning for the big picture 
• Explore the Resource Center documentation for setup and capabilities 
• Watch the configuration webinar for a practical walkthrough 

Join the discussion 

What are you trying to solve with AI in your organization? 

• Are you evaluating whether LAM Agentic Virtual Agent is the right fit? 
• Do you have a use case in mind but aren't sure where it fits? 
• Curious how others are getting started? 

Drop your questions below, our experts are here to help. 

Genesys Education is here, along with some experts from the Genesys Product Team to share insights on LAM powered Agentic Virtual Agents, a new approach to virtual self-service that can reason, take action, and move customers toward outcomes, not just respond to questions.  

Expert: Mitchell Mason - Principal Product Manager, Virtual Agent

If you're evaluating AI investments, wondering if this is right for your business, or thinking about how to get started, read on.  

Q1: What's the #1 thing new users should understand about LAM Agentic Virtual Agents? 

LAM Agentic Virtual Agents aren't just more conversational bots, they're designed to work toward an outcome. Instead of following a fixed flow or just answering questions, they can: 

• Understand a customer's goal
• Decide what steps are needed 
• Use tools and knowledge to take action 

That shift, from responding to reasoning + acting, is what makes them different. 

Q2: When are LAM Agentic Virtual Agents the right fit? 

LAM Agentic Virtual Agents are a strong fit when a customer interaction requires both conversation and action, especially when the path isn't predictable. 

Good examples: 

• Checking eligibility, then applying a change 
• Pulling info from multiple systems, then acting on it 
• Handling requests that evolve mid-conversation 

Not every use case needs this level of autonomy. 

• If you mainly need Q&A from knowledge, an LLM-powered virtual agent may be enough 
• If you need strict, repeatable workflows, AI Guides may be a better fit 

The key question: Do you need the system to decide and act, or just respond? 

Q3: What should decision makers think about before investing? 

Start with the problem, not the technology. 

LAM Agentic Virtual Agents work best when: 

• The customer goal is valuable to automate 
• The journey has enough variability that rigid flows break down 
• You can clearly define what the agent should and shouldn't do 

Also think about: 

• What systems or data it needs access to 
• What actions should stay human-led 
• How you'll measure success (containment, speed, effort, etc.) 

A focused first use case usually beats a broad one. 

Q4: What do admins and architects need to plan for? 

The experience may feel simple, but good setup still matters. 

In Genesys Cloud, you'll configure: 

• The agent's role and guidelines that keep it aligned to policy 
• The tools it can use (actions across systems) 
• The knowledge it can draw from 
• The guardrails that prevent unwanted behavior 

Then you connect it into Architect using a Virtual Agent flow where you can also determine additional aspects such as what information can be shared 

So, the work shifts from building hundreds of steps to making smart decisions about scope, access, and boundaries. 

Q5: What are some key considerations when starting with LAM Powered Agentic Virtual Agents?

Early decisions and focus areas that can shape your success:

• Start with a focused, high-value journey rather than casting too wide a net early on
• Invest in the right tools and knowledge to fully enable what the agent can accomplish
• Design with clear outcomes in mind, not just the conversation itself
• Evaluate the fit upfront to ensure LAM Agentic Virtual Agents are the right approach versus a simpler solution

Q6: Do you have a favorite tip for getting started? 

Yes: design around a customer outcome, not the feature set. Pick one journey where: 

• Customers struggle today 
• Multiple steps or systems are involved 
• Better reasoning + action would clearly help 

Start there, test, refine, and then expand. 

Q7: Where can learners go for more info? 

• Start with the LAM Agentic Virtual Agent Introduction eLearning for the big picture 
• Explore the Resource Center documentation for setup and capabilities 
• Watch the configuration webinar for a practical walkthrough 

Join the discussion 

What are you trying to solve with AI in your organization? 

• Are you evaluating whether LAM Agentic Virtual Agent is the right fit? 
• Do you have a use case in mind but aren't sure where it fits? 
• Curious how others are getting started? 

Drop your questions below, our experts are here to help. 

Genesys Education is here, along with some experts from the Genesys Product Team to share insights on LAM powered Agentic Virtual Agents, a new approach to virtual self-service that can reason, take action, and move customers toward outcomes, not just respond to questions.  

Expert: Mitchell Mason - Principal Product Manager, Virtual Agent

If you're evaluating AI investments, wondering if this is right for your business, or thinking about how to get started, read on.  

Q1: What's the #1 thing new users should understand about LAM Agentic Virtual Agents? 

LAM Agentic Virtual Agents aren't just more conversational bots, they're designed to work toward an outcome. Instead of following a fixed flow or just answering questions, they can: 

• Understand a customer's goal
• Decide what steps are needed 
• Use tools and knowledge to take action 

That shift, from responding to reasoning + acting, is what makes them different. 

Q2: When are LAM Agentic Virtual Agents the right fit? 

LAM Agentic Virtual Agents are a strong fit when a customer interaction requires both conversation and action, especially when the path isn't predictable. 

Good examples: 

• Checking eligibility, then applying a change 
• Pulling info from multiple systems, then acting on it 
• Handling requests that evolve mid-conversation 

Not every use case needs this level of autonomy. 

• If you mainly need Q&A from knowledge, an LLM-powered virtual agent may be enough 
• If you need strict, repeatable workflows, AI Guides may be a better fit 

The key question: Do you need the system to decide and act, or just respond? 

Q3: What should decision makers think about before investing? 

Start with the problem, not the technology. 

LAM Agentic Virtual Agents work best when: 

• The customer goal is valuable to automate 
• The journey has enough variability that rigid flows break down 
• You can clearly define what the agent should and shouldn't do 

Also think about: 

• What systems or data it needs access to 
• What actions should stay human-led 
• How you'll measure success (containment, speed, effort, etc.) 

A focused first use case usually beats a broad one. 

Q4: What do admins and architects need to plan for? 

The experience may feel simple, but good setup still matters. 

In Genesys Cloud, you'll configure: 

• The agent's role and guidelines that keep it aligned to policy 
• The tools it can use (actions across systems) 
• The knowledge it can draw from 
• The guardrails that prevent unwanted behavior 

Then you connect it into Architect using a Virtual Agent flow where you can also determine additional aspects such as what information can be shared 

So, the work shifts from building hundreds of steps to making smart decisions about scope, access, and boundaries. 

Q5: What are some key considerations when starting with LAM Powered Agentic Virtual Agents?

Early decisions and focus areas that can shape your success:

• Start with a focused, high-value journey rather than casting too wide a net early on
• Invest in the right tools and knowledge to fully enable what the agent can accomplish
• Design with clear outcomes in mind, not just the conversation itself
• Evaluate the fit upfront to ensure LAM Agentic Virtual Agents are the right approach versus a simpler solution

Q6: Do you have a favorite tip for getting started? 

Yes: design around a customer outcome, not the feature set. Pick one journey where: 

• Customers struggle today 
• Multiple steps or systems are involved 
• Better reasoning + action would clearly help 

Start there, test, refine, and then expand. 

Q7: Where can learners go for more info? 

• Start with the LAM Agentic Virtual Agent Introduction eLearning for the big picture 
• Explore the Resource Center documentation for setup and capabilities 
• Watch the configuration webinar for a practical walkthrough 

Join the discussion 

What are you trying to solve with AI in your organization? 

• Are you evaluating whether LAM Agentic Virtual Agent is the right fit? 
• Do you have a use case in mind but aren't sure where it fits? 
• Curious how others are getting started? 

Drop your questions below, our experts are here to help. 

Genesys Education is here, along with some experts from the Genesys Product Team to share insights on LAM powered Agentic Virtual Agents, a new approach to virtual self-service that can reason, take action, and move customers toward outcomes, not just respond to questions.  

Expert: Mitchell Mason - Principal Product Manager, Virtual Agent

If you're evaluating AI investments, wondering if this is right for your business, or thinking about how to get started, read on.  

Q1: What's the #1 thing new users should understand about LAM Agentic Virtual Agents? 

LAM Agentic Virtual Agents aren't just more conversational bots, they're designed to work toward an outcome. Instead of following a fixed flow or just answering questions, they can: 

• Understand a customer's goal
• Decide what steps are needed 
• Use tools and knowledge to take action 

That shift, from responding to reasoning + acting, is what makes them different. 

Q2: When are LAM Agentic Virtual Agents the right fit? 

LAM Agentic Virtual Agents are a strong fit when a customer interaction requires both conversation and action, especially when the path isn't predictable. 

Good examples: 

• Checking eligibility, then applying a change 
• Pulling info from multiple systems, then acting on it 
• Handling requests that evolve mid-conversation 

Not every use case needs this level of autonomy. 

• If you mainly need Q&A from knowledge, an LLM-powered virtual agent may be enough 
• If you need strict, repeatable workflows, AI Guides may be a better fit 

The key question: Do you need the system to decide and act, or just respond? 

Q3: What should decision makers think about before investing? 

Start with the problem, not the technology. 

LAM Agentic Virtual Agents work best when: 

• The customer goal is valuable to automate 
• The journey has enough variability that rigid flows break down 
• You can clearly define what the agent should and shouldn't do 

Also think about: 

• What systems or data it needs access to 
• What actions should stay human-led 
• How you'll measure success (containment, speed, effort, etc.) 

A focused first use case usually beats a broad one. 

Q4: What do admins and architects need to plan for? 

The experience may feel simple, but good setup still matters. 

In Genesys Cloud, you'll configure: 

• The agent's role and guidelines that keep it aligned to policy 
• The tools it can use (actions across systems) 
• The knowledge it can draw from 
• The guardrails that prevent unwanted behavior 

Then you connect it into Architect using a Virtual Agent flow where you can also determine additional aspects such as what information can be shared 

So, the work shifts from building hundreds of steps to making smart decisions about scope, access, and boundaries. 

Q5: What are some key considerations when starting with LAM Powered Agentic Virtual Agents?

Early decisions and focus areas that can shape your success:

• Start with a focused, high-value journey rather than casting too wide a net early on
• Invest in the right tools and knowledge to fully enable what the agent can accomplish
• Design with clear outcomes in mind, not just the conversation itself
• Evaluate the fit upfront to ensure LAM Agentic Virtual Agents are the right approach versus a simpler solution

Q6: Do you have a favorite tip for getting started? 

Yes: design around a customer outcome, not the feature set. Pick one journey where: 

• Customers struggle today 
• Multiple steps or systems are involved 
• Better reasoning + action would clearly help 

Start there, test, refine, and then expand. 

Q7: Where can learners go for more info? 

• Start with the LAM Agentic Virtual Agent Introduction eLearning for the big picture 
• Explore the Resource Center documentation for setup and capabilities 
• Watch the configuration webinar for a practical walkthrough 

Join the discussion 

What are you trying to solve with AI in your organization? 

• Are you evaluating whether LAM Agentic Virtual Agent is the right fit? 
• Do you have a use case in mind but aren't sure where it fits? 
• Curious how others are getting started? 

Drop your questions below, our experts are here to help. 

Genesys Education is here, along with some experts from the Genesys Product Team to share insights on LAM powered Agentic Virtual Agents, a new approach to virtual self-service that can reason, take action, and move customers toward outcomes, not just respond to questions.  

Expert: Mitchell Mason - Principal Product Manager, Virtual Agent

If you're evaluating AI investments, wondering if this is right for your business, or thinking about how to get started, read on.  

Q1: What's the #1 thing new users should understand about LAM Agentic Virtual Agents? 

LAM Agentic Virtual Agents aren't just more conversational bots, they're designed to work toward an outcome. Instead of following a fixed flow or just answering questions, they can: 

• Understand a customer's goal
• Decide what steps are needed 
• Use tools and knowledge to take action 

That shift, from responding to reasoning + acting, is what makes them different. 

Q2: When are LAM Agentic Virtual Agents the right fit? 

LAM Agentic Virtual Agents are a strong fit when a customer interaction requires both conversation and action, especially when the path isn't predictable. 

Good examples: 

• Checking eligibility, then applying a change 
• Pulling info from multiple systems, then acting on it 
• Handling requests that evolve mid-conversation 

Not every use case needs this level of autonomy. 

• If you mainly need Q&A from knowledge, an LLM-powered virtual agent may be enough 
• If you need strict, repeatable workflows, AI Guides may be a better fit 

The key question: Do you need the system to decide and act, or just respond? 

Q3: What should decision makers think about before investing? 

Start with the problem, not the technology. 

LAM Agentic Virtual Agents work best when: 

• The customer goal is valuable to automate 
• The journey has enough variability that rigid flows break down 
• You can clearly define what the agent should and shouldn't do 

Also think about: 

• What systems or data it needs access to 
• What actions should stay human-led 
• How you'll measure success (containment, speed, effort, etc.) 

A focused first use case usually beats a broad one. 

Q4: What do admins and architects need to plan for? 

The experience may feel simple, but good setup still matters. 

In Genesys Cloud, you'll configure: 

• The agent's role and guidelines that keep it aligned to policy 
• The tools it can use (actions across systems) 
• The knowledge it can draw from 
• The guardrails that prevent unwanted behavior 

Then you connect it into Architect using a Virtual Agent flow where you can also determine additional aspects such as what information can be shared 

So, the work shifts from building hundreds of steps to making smart decisions about scope, access, and boundaries. 

Q5: What are some key considerations when starting with LAM Powered Agentic Virtual Agents?

Early decisions and focus areas that can shape your success:

• Start with a focused, high-value journey rather than casting too wide a net early on
• Invest in the right tools and knowledge to fully enable what the agent can accomplish
• Design with clear outcomes in mind, not just the conversation itself
• Evaluate the fit upfront to ensure LAM Agentic Virtual Agents are the right approach versus a simpler solution

Q6: Do you have a favorite tip for getting started? 

Yes: design around a customer outcome, not the feature set. Pick one journey where: 

• Customers struggle today 
• Multiple steps or systems are involved 
• Better reasoning + action would clearly help 

Start there, test, refine, and then expand. 

Q7: Where can learners go for more info? 

• Start with the LAM Agentic Virtual Agent Introduction eLearning for the big picture 
• Explore the Resource Center documentation for setup and capabilities 
• Watch the configuration webinar for a practical walkthrough 

Join the discussion 

What are you trying to solve with AI in your organization? 

• Are you evaluating whether LAM Agentic Virtual Agent is the right fit? 
• Do you have a use case in mind but aren't sure where it fits? 
• Curious how others are getting started? 

Drop your questions below, our experts are here to help. 

Genesys Education is here, along with some experts from the Genesys Product Team to share insights on LAM powered Agentic Virtual Agents, a new approach to virtual self-service that can reason, take action, and move customers toward outcomes, not just respond to questions.  

Expert: Mitchell Mason - Principal Product Manager, Virtual Agent

If you're evaluating AI investments, wondering if this is right for your business, or thinking about how to get started, read on.  

Q1: What's the #1 thing new users should understand about LAM Agentic Virtual Agents? 

LAM Agentic Virtual Agents aren't just more conversational bots, they're designed to work toward an outcome. Instead of following a fixed flow or just answering questions, they can: 

• Understand a customer's goal
• Decide what steps are needed 
• Use tools and knowledge to take action 

That shift, from responding to reasoning + acting, is what makes them different. 

Q2: When are LAM Agentic Virtual Agents the right fit? 

LAM Agentic Virtual Agents are a strong fit when a customer interaction requires both conversation and action, especially when the path isn't predictable. 

Good examples: 

• Checking eligibility, then applying a change 
• Pulling info from multiple systems, then acting on it 
• Handling requests that evolve mid-conversation 

Not every use case needs this level of autonomy. 

• If you mainly need Q&A from knowledge, an LLM-powered virtual agent may be enough 
• If you need strict, repeatable workflows, AI Guides may be a better fit 

The key question: Do you need the system to decide and act, or just respond? 

Q3: What should decision makers think about before investing? 

Start with the problem, not the technology. 

LAM Agentic Virtual Agents work best when: 

• The customer goal is valuable to automate 
• The journey has enough variability that rigid flows break down 
• You can clearly define what the agent should and shouldn't do 

Also think about: 

• What systems or data it needs access to 
• What actions should stay human-led 
• How you'll measure success (containment, speed, effort, etc.) 

A focused first use case usually beats a broad one. 

Q4: What do admins and architects need to plan for? 

The experience may feel simple, but good setup still matters. 

In Genesys Cloud, you'll configure: 

• The agent's role and guidelines that keep it aligned to policy 
• The tools it can use (actions across systems) 
• The knowledge it can draw from 
• The guardrails that prevent unwanted behavior 

Then you connect it into Architect using a Virtual Agent flow where you can also determine additional aspects such as what information can be shared 

So, the work shifts from building hundreds of steps to making smart decisions about scope, access, and boundaries. 

Q5: What are some key considerations when starting with LAM Powered Agentic Virtual Agents?

Early decisions and focus areas that can shape your success:

• Start with a focused, high-value journey rather than casting too wide a net early on
• Invest in the right tools and knowledge to fully enable what the agent can accomplish
• Design with clear outcomes in mind, not just the conversation itself
• Evaluate the fit upfront to ensure LAM Agentic Virtual Agents are the right approach versus a simpler solution

Q6: Do you have a favorite tip for getting started? 

Yes: design around a customer outcome, not the feature set. Pick one journey where: 

• Customers struggle today 
• Multiple steps or systems are involved 
• Better reasoning + action would clearly help 

Start there, test, refine, and then expand. 

Q7: Where can learners go for more info? 

• Start with the LAM Agentic Virtual Agent Introduction eLearning for the big picture 
• Explore the Resource Center documentation for setup and capabilities 
• Watch the configuration webinar for a practical walkthrough 

Join the discussion 

What are you trying to solve with AI in your organization? 

• Are you evaluating whether LAM Agentic Virtual Agent is the right fit? 
• Do you have a use case in mind but aren't sure where it fits? 
• Curious how others are getting started? 

Drop your questions below, our experts are here to help. 

Genesys Education is here, along with some experts from the Genesys Product Team to share insights on LAM powered Agentic Virtual Agents, a new approach to virtual self-service that can reason, take action, and move customers toward outcomes, not just respond to questions.  

Expert: Mitchell Mason - Principal Product Manager, Virtual Agent

If you're evaluating AI investments, wondering if this is right for your business, or thinking about how to get started, read on.  

Q1: What's the #1 thing new users should understand about LAM Agentic Virtual Agents? 

LAM Agentic Virtual Agents aren't just more conversational bots, they're designed to work toward an outcome. Instead of following a fixed flow or just answering questions, they can: 

• Understand a customer's goal
• Decide what steps are needed 
• Use tools and knowledge to take action 

That shift, from responding to reasoning + acting, is what makes them different. 

Q2: When are LAM Agentic Virtual Agents the right fit? 

LAM Agentic Virtual Agents are a strong fit when a customer interaction requires both conversation and action, especially when the path isn't predictable. 

Good examples: 

• Checking eligibility, then applying a change 
• Pulling info from multiple systems, then acting on it 
• Handling requests that evolve mid-conversation 

Not every use case needs this level of autonomy. 

• If you mainly need Q&A from knowledge, an LLM-powered virtual agent may be enough 
• If you need strict, repeatable workflows, AI Guides may be a better fit 

The key question: Do you need the system to decide and act, or just respond? 

Q3: What should decision makers think about before investing? 

Start with the problem, not the technology. 

LAM Agentic Virtual Agents work best when: 

• The customer goal is valuable to automate 
• The journey has enough variability that rigid flows break down 
• You can clearly define what the agent should and shouldn't do 

Also think about: 

• What systems or data it needs access to 
• What actions should stay human-led 
• How you'll measure success (containment, speed, effort, etc.) 

A focused first use case usually beats a broad one. 

Q4: What do admins and architects need to plan for? 

The experience may feel simple, but good setup still matters. 

In Genesys Cloud, you'll configure: 

• The agent's role and guidelines that keep it aligned to policy 
• The tools it can use (actions across systems) 
• The knowledge it can draw from 
• The guardrails that prevent unwanted behavior 

Then you connect it into Architect using a Virtual Agent flow where you can also determine additional aspects such as what information can be shared 

So, the work shifts from building hundreds of steps to making smart decisions about scope, access, and boundaries. 

Q5: What are some key considerations when starting with LAM Powered Agentic Virtual Agents?

Early decisions and focus areas that can shape your success:

• Start with a focused, high-value journey rather than casting too wide a net early on
• Invest in the right tools and knowledge to fully enable what the agent can accomplish
• Design with clear outcomes in mind, not just the conversation itself
• Evaluate the fit upfront to ensure LAM Agentic Virtual Agents are the right approach versus a simpler solution

Q6: Do you have a favorite tip for getting started? 

Yes: design around a customer outcome, not the feature set. Pick one journey where: 

• Customers struggle today 
• Multiple steps or systems are involved 
• Better reasoning + action would clearly help 

Start there, test, refine, and then expand. 

Q7: Where can learners go for more info? 

• Start with the LAM Agentic Virtual Agent Introduction eLearning for the big picture 
• Explore the Resource Center documentation for setup and capabilities 
• Watch the configuration webinar for a practical walkthrough 

Join the discussion 

What are you trying to solve with AI in your organization? 

• Are you evaluating whether LAM Agentic Virtual Agent is the right fit? 
• Do you have a use case in mind but aren't sure where it fits? 
• Curious how others are getting started? 

Drop your questions below, our experts are here to help. 

Genesys Education is here, along with some experts from the Genesys Product Team to share insights on LAM powered Agentic Virtual Agents, a new approach to virtual self-service that can reason, take action, and move customers toward outcomes, not just respond to questions.  

Expert: Mitchell Mason - Principal Product Manager, Virtual Agent

If you're evaluating AI investments, wondering if this is right for your business, or thinking about how to get started, read on.  

Q1: What's the #1 thing new users should understand about LAM Agentic Virtual Agents? 

LAM Agentic Virtual Agents aren't just more conversational bots, they're designed to work toward an outcome. Instead of following a fixed flow or just answering questions, they can: 

• Understand a customer's goal
• Decide what steps are needed 
• Use tools and knowledge to take action 

That shift, from responding to reasoning + acting, is what makes them different. 

Q2: When are LAM Agentic Virtual Agents the right fit? 

LAM Agentic Virtual Agents are a strong fit when a customer interaction requires both conversation and action, especially when the path isn't predictable. 

Good examples: 

• Checking eligibility, then applying a change 
• Pulling info from multiple systems, then acting on it 
• Handling requests that evolve mid-conversation 

Not every use case needs this level of autonomy. 

• If you mainly need Q&A from knowledge, an LLM-powered virtual agent may be enough 
• If you need strict, repeatable workflows, AI Guides may be a better fit 

The key question: Do you need the system to decide and act, or just respond? 

Q3: What should decision makers think about before investing? 

Start with the problem, not the technology. 

LAM Agentic Virtual Agents work best when: 

• The customer goal is valuable to automate 
• The journey has enough variability that rigid flows break down 
• You can clearly define what the agent should and shouldn't do 

Also think about: 

• What systems or data it needs access to 
• What actions should stay human-led 
• How you'll measure success (containment, speed, effort, etc.) 

A focused first use case usually beats a broad one. 

Q4: What do admins and architects need to plan for? 

The experience may feel simple, but good setup still matters. 

In Genesys Cloud, you'll configure: 

• The agent's role and guidelines that keep it aligned to policy 
• The tools it can use (actions across systems) 
• The knowledge it can draw from 
• The guardrails that prevent unwanted behavior 

Then you connect it into Architect using a Virtual Agent flow where you can also determine additional aspects such as what information can be shared 

So, the work shifts from building hundreds of steps to making smart decisions about scope, access, and boundaries. 

Q5: What are some key considerations when starting with LAM Powered Agentic Virtual Agents?

Early decisions and focus areas that can shape your success:

• Start with a focused, high-value journey rather than casting too wide a net early on
• Invest in the right tools and knowledge to fully enable what the agent can accomplish
• Design with clear outcomes in mind, not just the conversation itself
• Evaluate the fit upfront to ensure LAM Agentic Virtual Agents are the right approach versus a simpler solution

Q6: Do you have a favorite tip for getting started? 

Yes: design around a customer outcome, not the feature set. Pick one journey where: 

• Customers struggle today 
• Multiple steps or systems are involved 
• Better reasoning + action would clearly help 

Start there, test, refine, and then expand. 

Q7: Where can learners go for more info? 

• Start with the LAM Agentic Virtual Agent Introduction eLearning for the big picture 
• Explore the Resource Center documentation for setup and capabilities 
• Watch the configuration webinar for a practical walkthrough 

Join the discussion 

What are you trying to solve with AI in your organization? 

• Are you evaluating whether LAM Agentic Virtual Agent is the right fit? 
• Do you have a use case in mind but aren't sure where it fits? 
• Curious how others are getting started? 

Drop your questions below, our experts are here to help. 

Genesys Education is here, along with some experts from the Genesys Product Team to share insights on LAM powered Agentic Virtual Agents, a new approach to virtual self-service that can reason, take action, and move customers toward outcomes, not just respond to questions.  

Expert: Mitchell Mason - Principal Product Manager, Virtual Agent

If you're evaluating AI investments, wondering if this is right for your business, or thinking about how to get started, read on.  

Q1: What's the #1 thing new users should understand about LAM Agentic Virtual Agents? 

LAM Agentic Virtual Agents aren't just more conversational bots, they're designed to work toward an outcome. Instead of following a fixed flow or just answering questions, they can: 

• Understand a customer's goal
• Decide what steps are needed 
• Use tools and knowledge to take action 

That shift, from responding to reasoning + acting, is what makes them different. 

Q2: When are LAM Agentic Virtual Agents the right fit? 

LAM Agentic Virtual Agents are a strong fit when a customer interaction requires both conversation and action, especially when the path isn't predictable. 

Good examples: 

• Checking eligibility, then applying a change 
• Pulling info from multiple systems, then acting on it 
• Handling requests that evolve mid-conversation 

Not every use case needs this level of autonomy. 

• If you mainly need Q&A from knowledge, an LLM-powered virtual agent may be enough 
• If you need strict, repeatable workflows, AI Guides may be a better fit 

The key question: Do you need the system to decide and act, or just respond? 

Q3: What should decision makers think about before investing? 

Start with the problem, not the technology. 

LAM Agentic Virtual Agents work best when: 

• The customer goal is valuable to automate 
• The journey has enough variability that rigid flows break down 
• You can clearly define what the agent should and shouldn't do 

Also think about: 

• What systems or data it needs access to 
• What actions should stay human-led 
• How you'll measure success (containment, speed, effort, etc.) 

A focused first use case usually beats a broad one. 

Q4: What do admins and architects need to plan for? 

The experience may feel simple, but good setup still matters. 

In Genesys Cloud, you'll configure: 

• The agent's role and guidelines that keep it aligned to policy 
• The tools it can use (actions across systems) 
• The knowledge it can draw from 
• The guardrails that prevent unwanted behavior 

Then you connect it into Architect using a Virtual Agent flow where you can also determine additional aspects such as what information can be shared 

So, the work shifts from building hundreds of steps to making smart decisions about scope, access, and boundaries. 

Q5: What are some key considerations when starting with LAM Powered Agentic Virtual Agents?

Early decisions and focus areas that can shape your success:

• Start with a focused, high-value journey rather than casting too wide a net early on
• Invest in the right tools and knowledge to fully enable what the agent can accomplish
• Design with clear outcomes in mind, not just the conversation itself
• Evaluate the fit upfront to ensure LAM Agentic Virtual Agents are the right approach versus a simpler solution

Q6: Do you have a favorite tip for getting started? 

Yes: design around a customer outcome, not the feature set. Pick one journey where: 

• Customers struggle today 
• Multiple steps or systems are involved 
• Better reasoning + action would clearly help 

Start there, test, refine, and then expand. 

Q7: Where can learners go for more info? 

• Start with the LAM Agentic Virtual Agent Introduction eLearning for the big picture 
• Explore the Resource Center documentation for setup and capabilities 
• Watch the configuration webinar for a practical walkthrough 

Join the discussion 

What are you trying to solve with AI in your organization? 

• Are you evaluating whether LAM Agentic Virtual Agent is the right fit? 
• Do you have a use case in mind but aren't sure where it fits? 
• Curious how others are getting started? 

Drop your questions below, our experts are here to help. 

Genesys Education is here, along with some experts from the Genesys Product Team to share insights on LAM powered Agentic Virtual Agents, a new approach to virtual self-service that can reason, take action, and move customers toward outcomes, not just respond to questions.  

Expert: Mitchell Mason - Principal Product Manager, Virtual Agent

If you're evaluating AI investments, wondering if this is right for your business, or thinking about how to get started, read on.  

Q1: What's the #1 thing new users should understand about LAM Agentic Virtual Agents? 

LAM Agentic Virtual Agents aren't just more conversational bots, they're designed to work toward an outcome. Instead of following a fixed flow or just answering questions, they can: 

• Understand a customer's goal
• Decide what steps are needed 
• Use tools and knowledge to take action 

That shift, from responding to reasoning + acting, is what makes them different. 

Q2: When are LAM Agentic Virtual Agents the right fit? 

LAM Agentic Virtual Agents are a strong fit when a customer interaction requires both conversation and action, especially when the path isn't predictable. 

Good examples: 

• Checking eligibility, then applying a change 
• Pulling info from multiple systems, then acting on it 
• Handling requests that evolve mid-conversation 

Not every use case needs this level of autonomy. 

• If you mainly need Q&A from knowledge, an LLM-powered virtual agent may be enough 
• If you need strict, repeatable workflows, AI Guides may be a better fit 

The key question: Do you need the system to decide and act, or just respond? 

Q3: What should decision makers think about before investing? 

Start with the problem, not the technology. 

LAM Agentic Virtual Agents work best when: 

• The customer goal is valuable to automate 
• The journey has enough variability that rigid flows break down 
• You can clearly define what the agent should and shouldn't do 

Also think about: 

• What systems or data it needs access to 
• What actions should stay human-led 
• How you'll measure success (containment, speed, effort, etc.) 

A focused first use case usually beats a broad one. 

Q4: What do admins and architects need to plan for? 

The experience may feel simple, but good setup still matters. 

In Genesys Cloud, you'll configure: 

• The agent's role and guidelines that keep it aligned to policy 
• The tools it can use (actions across systems) 
• The knowledge it can draw from 
• The guardrails that prevent unwanted behavior 

Then you connect it into Architect using a Virtual Agent flow where you can also determine additional aspects such as what information can be shared 

So, the work shifts from building hundreds of steps to making smart decisions about scope, access, and boundaries. 

Q5: What are some key considerations when starting with LAM Powered Agentic Virtual Agents?

Early decisions and focus areas that can shape your success:

• Start with a focused, high-value journey rather than casting too wide a net early on
• Invest in the right tools and knowledge to fully enable what the agent can accomplish
• Design with clear outcomes in mind, not just the conversation itself
• Evaluate the fit upfront to ensure LAM Agentic Virtual Agents are the right approach versus a simpler solution

Q6: Do you have a favorite tip for getting started? 

Yes: design around a customer outcome, not the feature set. Pick one journey where: 

• Customers struggle today 
• Multiple steps or systems are involved 
• Better reasoning + action would clearly help 

Start there, test, refine, and then expand. 

Q7: Where can learners go for more info? 

• Start with the LAM Agentic Virtual Agent Introduction eLearning for the big picture 
• Explore the Resource Center documentation for setup and capabilities 
• Watch the configuration webinar for a practical walkthrough 

Join the discussion 

What are you trying to solve with AI in your organization? 

• Are you evaluating whether LAM Agentic Virtual Agent is the right fit? 
• Do you have a use case in mind but aren't sure where it fits? 
• Curious how others are getting started? 

Drop your questions below, our experts are here to help.