Build Your First AI Agent in One Hour

Good afternoon, everyone, and welcome to Women Applying AI, our session today, I'm very excited about. But first, I just want to welcome you and thank you for being a part of our community, which as you know, offers women a collaborative and safe space and environment to learn AI, learn by doing, grow your network, grow your possibilities and grow all of your knowledge and be ready for an AI future. And not just ready for an AI future, but really what we want to do is empower women to lead and shape the AI future because we know that if we're not involved, it's not going to be as cool as it could be, absolutely. So my name is Aileen McDonough, I am a founding member of WAAI and also head of programs.

And I am very excited to welcome you to this amazing session. I know it's going to be so great with Jacqueline Twillie, build your first AI agent in one hour. We toyed with what to call this. And I was almost saying like build your first AI chief of staff in one hour, but there's a lot of different agents we can build. So we wanted to keep it open to that. So without further ado, I'm going to hand it off to Jacqueline, I'm going to stop sharing and she will be sharing. And I'll give you a few logistics while she gets up, while she gets going.

So this is very much an interactive session today, but I did want to let you know that if you have any questions about the build, first of all, just want to confirm that everybody got the Gumloop free trial, because that will be necessary for building along with Jacqueline today. There is a link in the chat, so feel free to grab that quickly and download. Because this is interactive, we welcome questions about the build during, but if you have more general questions, please just put them in the chat and feel free to upvote questions if you have a similar question, and we will get to that at the end as part of a Q&A from Jacqueline.

So take it away, Jacqueline. Well, hello, everyone. I am so excited about today and building our first AI agents. We're actually going to build the agent real time. So if you haven't had a moment, please do use the link in the chat to create your free trial account. A best practice is go ahead and also put a note in your phone to cancel in 12 days if you're only doing a trial run. That way you don't get a surprise charge in 14 days. So I like to start off with that. I am based in New York. I have a non-technical background. Over the summer, I started teaching a group of friends to build agents in one hour.

Since then, over 300 people have joined me, and I am most excited to be with WAAI today for us to build our agents. So as we get started, I just want you to come in agreement with me on two things, okay? One, if you do have a question about the build, please ask the question. My only request is that you don't apologize for asking a question. Be comfortable asking that question. The second thing is if you find something useful and helpful, don't keep the information to yourself. The way that we close the AI fluency gap is by sharing what we learned with others. So if I can get you to give me a thumbs up in the chat that you agree to ask a question but don't apologize if you have one.

And two, share something that you find useful. That makes my day. I like us all to be in agreement with that. No gatekeeping. Fantastic. Great. So I'm sharing with you my screen right now. I'm going to give you all access to my digital presentation. It walks us through step by step, but what I'm going to do for us now is we're going to slide into Gumloop. So what I would like you to do is go to the home page of Gumloop, and what we're going to do is we're going to walk through building our agents. So if you have a routine that you repeat frequently, you can customize this agent to take on that repetitive task.

But in general, we're going to start off with a chief of staff agent today. That's something most people can relate to. But if you came into this session saying, Jacqueline, I know exactly what I'd use an AI agent for, feel free to deviate slightly and you'll see where you can make those deviations to get your agent up and going. First question is, what the heck is an AI agent? So an AI agent at the non-technical definition, think of it as a chat bot that you give instructions as well as access to certain systems, perhaps calendar, email, Slack, Salesforce, et cetera. So once an agent has access to instructions as well as systems, then they're able to complete the workflow for you.

That is the non-technical version for the purpose of we only have an hour today. So once we are bringing on our agent, I'd like to think of this as hiring a really competent new hire, and I'm giving it an onboarding. If I'm onboarding a new hire and I give them great instructions and access to systems, chances are I'm going to see pretty good results. However, if I'm doing an onboarding and I leave out critical information and I restrict access and I don't provide access to systems, then that new hire isn't going to achieve as much. So just keep that in the back of mind as we go through the process and we start building.

There are two common questions that come up that I just want to address before we start building. And the first one is, what about privacy? And the second one is, what about security? So for those who may have those questions, you'll see in the chat, I've just resurfaced Gumloop's policy on privacy and security. All right. Now the next big question that I get is, why isn't this only a ChatGPT or Claude? And guess what, these models can run on Claude or ChatGPT for building on Gumloop, which is a no code agent orchestrator. So many of you may recall a couple of months ago, the government said, Claude, you need to roll back your fable five model.

We think it's too powerful and we don't want people to touch it yet. One of the great things about a system like Gumloop is you can move your model for your agent from one large language model to the next in one click. And if one model becomes too expensive, you can actually slide over to the next model easily. So that's why we're not focused on just ChatGPT or Claude. And to borrow a quote from one of my amazing WAAI friends, Erica Rooney, we would never say when it comes to driving a car, we can only drive a Toyota. So when it comes to AI tools, we want to be able to get behind the wheel and be fluent and experiment. So that's enough of the intro.

Who's ready to build your first agent? Give me a thumbs up in the chat. Let me know if you set those accounts up and we're ready to go. I see a couple of thumbs, I need, okay, fantastic, you all are great. This is amazing. My screen may look a little different from yours. I'm currently running over 30 agents inside of Gumloop. Today we're going to start building one. To get started, we'll go to the home screen of Gumloop. You should see the Gumloop logo in the top left corner. I'm actually hovering over it right now. There's also a word that says home. Beneath home, you'll see agents. When you're hovering over agent, if you slide over, there's a magnifying glass icon.

It's slightly to the right of the word agent. Next to that is a plus icon. Please click on the plus icon. The plus icon will bring you to the screen that you see on my session now. So if you're looking at my screen, you'll see something that says, what should this agent take care of, question and mark. So, this is where we start building our agent. There are three rules that I like to follow as a best practice for setting up an AI agent. Number one, give the agent a persona. So for those of you who are building a chief of staff agent alongside us today, we're going to first tell our agent, you are my chief of staff agent.

So go ahead and type that in the screen. I actually have on some fancy nails today, so my typing is a little bit slower than usual. So bear with me. So we give our agents a persona. The next thing that we want to do is we want to tell the agent what its goals are. And this is for any agent, persona, goals, and deliverables. So for this next part, what we're going to do is we're going to use the dictate feature on our screen. Inside of Gumloop, there are two ways to set up an agent. We can set it up manually, but the quickest way to do it is to use the system that they have.

So using this chat feature to set it up. You may notice on this screen, there's a few drop downs, data analytics, et cetera, all the way down to sales outreach. Gumloop has made it really easy for non-technical professionals and teams to get started with an agent to prompt. So if you're looking at this list on the screen and you're thinking, actually, instead of a chief of staff agent, I want a marketing content agent, go ahead and select that option. You're more than welcome to do that. I'm going to walk you through how we customize this. So as a best practice when setting up an agent similar to how we onboard a new hire, we need to give really good instructions, contacts, and access to systems.

So what I typically do is I give my agent the persona. The next thing I'll tell my agent to do is, "Hey, these are the goals that I want you to achieve." Now, there's two ways that we can do this. We can type it all out or we can speak into it. I tend to find that speaking into the microphone for one to two minutes gives my system enough information to start working. So what we're going to do next is we'll mute ourselves on Zoom. So let's just double check before we start this. Make sure your Zoom is on mute. Once we verify that Zoom is on mute, we'll click the microphone feature. I'm going to do a demonstration on how to give your agent goals as well as deliverables.

My chief of staff agent, I'd like you to help me to stay on top of my quarterly goals. And my quarterly goals are to maintain great relationships with the Women Applying AI community. So I'd like to follow up on Slack messages, emails, and send useful information through Slack. Additionally, I want to be able to manage my calendar and not miss anything and not miss replying to critical emails at the end of day. I also want to ensure that you are, as my chief of staff agent, providing a dashboard so that I can visually see things that I need to take action on right away and things that maybe do that I've let slip through the cracks.

Be proactive and make smart decisions. And also, I shouldn't have to say this, but I will be honest with me. So I just demonstrated with you how I use the dictate feature inside of the window. What I'd like you to do is do something similar. Tell your agent what you wanted to assist you on. Now you'll notice I didn't tell my agent how to give me this information outside of a dashboard. So it'll ask me clarifying questions in a moment. What you now see on my screen is that the microphone actually translated into text. I probably would have only typed a sentence or two, especially with the fancy nails that I have on. But using voice mode, I got a lot done.

What I'd like you to do now is make sure you're on mute and go ahead and speak into your dictate feature. And I'll give you about two minutes to do so. When you finish, please put a thumbs up in the chat just so I can monitor where we are and we'll move forward from the next step after that. Do not press the up arrow yet. Only speak into your agent. And I'm here in the chat. If you have any questions, go ahead and speak into it.

Thank you.

All right, I'll give you about 30 more seconds. Just get to a natural stopping point. I see a couple of thumbs in the chat. Once you finish dictating the instructions, give me a thumbs up in the chat. Awesome. A few of us. I'll give it a few more seconds. Before we move to the next section, all right, I see a few more thumbs up coming through. Fantastic.

So continue to provide instructions and get to the point where you can get to a stopping point. What we'll do next is we're going to use the up arrow, which is the return enter feature. So what's going to happen next is our screen is going to change slightly. Do not be alarmed. This is a regular part of the build process. So the chat feature is there for us to get started in a really simple way, but we'll now see more controls once we hit the up arrow. Again, don't be alarmed. It is part of the normal process. What you'll see on your screen is that your agent starts to think. And you'll see that it actually dictates and transcribes what it's doing based on the instructions that you provided on my screen.

If you're following along on my screen, my agent is now asking for input. It's provided multiple choice questions for me. I can select from those options, but if none of those fit, I can use option D, which is other, and I can type in my response. So if you have questions on your screen, I'd like you to go ahead and answer those questions and hit submit. Your agent will begin to think again. While that's working, I want to call attention to this section that we see here. What's all going on? You'll see on the right hand side of the screen, if you're following along with my screen, I'm hovering over the word agent.

Next to that is a word triggers access and settings. So there are four words there that are clickable that you can click into. I'm going to click into settings. If you're following along, I'd like you also to click into settings. What I'm going to do is name my agent, COS, chief of staff agent. And I'm just going to do a demo agent and put WAAI because I actually have a chief of staff agent that I run day to day. I give my a participant hesitant to give Gumloop access to Gmail today. Can I just watch? Absolutely. And this is exactly why we pasted the links for the privacy policy as well as the security protocols so that you can see that a participant, it does not only use grok.

We're going to change that in one moment. I've actually put in a service ticket with the team for them to switch that so we can select our model up front. Great question. I see another question. Does it take past data into accounts or only emails that come in after the agent is set up? It depends on the level of permissions we give it. And we're going to be able to deep dive into that in one moment. Can we create a dummy Gmail account to practice with? Absolutely. That is one of the best practices in terms of building your comfort level with agents are actually a lot of people in Silicon Valley. They will set up an email account that is agents only. So you'll notice a lot of agents, personal AI agents, they will request access to email.

What a lot of folks are doing is setting up an agent account and only forwarding information that they want the agent to read. Yes, it is an extra step, but it's a great way to be comfortable with the level of access that you're giving to a system. So you can select a dummy account. What I recommend just starting out, most of us have an email account that we send like promotional emails, spam emails to just use that account to get started. Okay. All great questions. Continue to put them into the chat. So if you're looking at my screen, I've given my agent a name and this one has nothing to do with function. It's all about aesthetics. When I log into my dashboard, I come in to the icon section and select an emoji with color.

Again, this has nothing to do with function. It just gives me a little bit of flair when I come onto the screen. Once you give your agent a name, and again, this is under the settings section. If you're following my screen, I just clicked on the word settings and into the box where the agent Clever Ace name was. And I've given it COS Demo Agent WAAI, as well as I put a little flower plant there for growth, you know, growth mindset. We're learning something new. Once you do that, you do have to hit the save button. So inside of Gumloop, there are certain sections where you will advance forward and some screens you will remain on it unless you advance out of it. What happens here, when we hit save the word, the section grays out, so that confirms that the save has been successful.

The next thing that we're going to do is we're going to hover again over the word agent. If you're following along on my screen, it is on the right side of my screen. I'm clicking into agents below agent. You'll see model and there's a G, which is the Gumloop logo and auto. If you select auto in the dropdown menu, you'll see that there are all of the language, the large language lab models that are available that you can select from. So a moment ago, I said that if you want to switch to a cheaper model or if one model is no longer available, you can move in one click. This is where you would do that. So if you have a preferred model that you use, perhaps it's Gemini from Google, Claude from Anthropic, or ChatGPT from OpenAI, you can select it in the section here.

I typically build my new agents in Moonshot using the Kimi K3 model. It's one of the Chinese large language labs, Kimi K3 is one of my favorite in terms of setting up agents. The reason why I like the Kimi K3 model. It tends to give me great results and it doesn't use a lot of tokens inside of Gumloop. Right now we're all on a trial account, which gives us 20,000 credits to use. But I find that I get more bang for the buck when it comes to using the Kimi K3 model. If you're not familiar with the other models in the dropdown box, feel free to message me on LinkedIn. That's more than our hour will allow for us to do.

That can be a full hour on itself. So I'm happy to answer questions about models later. Like the model that you feel most comfortable with for today. And I would suggest if you're unsure safe space to start is one of the Google models. Gemini 3.1 flash or sorry, 3.8 flash or 3.7 flash, they tend to be really good general models. If you're unsure about which lab to go with, Google's doing a pretty good job right now, I'd say that's a safe starting point. All right. So select the model that you want. You'll notice here, there's a section that says add instructions for agents.

So if you're familiar with setting up agents, you can type in your own instructions. However, one of the benefits of a system like Gumloop, it's really adaptive. So the information that we provided in the first screen and then the questions that we've answered, it will create instructions for us. So we need the instructions box for the agent. You'll notice there's something that says edit own instructions with a blue toggle button that is on. What that does is it allows the agent to learn from your interaction inside of the chat feature. And you can start multiple chats. For example, if you click new task that opens another chat inside of this agent, what this blue button does is it allows your agent to learn how you're interacting with it and it automates updates.

So as a best practice, I have a 15 minute meeting that reoccurs every three weeks on my calendar where I jump in and I review my agent instructions so that as it's auto updating, I am still fully aware of what's happening. If you're using your agent daily, you may want to check yours every two weeks. Most of my agents, I have them running a couple of times a week. So a three week cadence works for me. Okay. But I just want to make you aware that if you keep the edit on own instructions on, you want to be able to manage and have oversight into what the agent is updating.

So please put a calendar reminder at some frequency for you to come back and check that. Okay. We need that. You'll see something called connectors. Now this is where I get super excited. I just clicked on the plus icon next to the word at, in the connector section. So what the heck are connectors? This is where we give access to calendar, email, Salesforce, et cetera, Slack, even connectors are what make agents work. This is what helps them to move from one part of a workflow to the next part. Earlier, I said, we're going to give our agent a persona. We give it goals.

We tell it deliverables. One of the cool things about agents that makes it a bit different from a chat bot is that it can advance to, all right, a participant, see you on the replay. It can advance to complete a task for us. So what that means is that if it's going to be able to decide, Hey, I need to check calendar. I need to check email and create an agenda for this meeting or send a follow-up recap, then it needs to have access to systems. So I'm going to share with you some of my favorite ones beyond email. If you feel comfortable setting up an email account and connecting your agent to it today, you can do that.

If you're not yet comfortable, just continue to follow along. I want to highlight for you three helpful connectors that are beyond calendar and email. The first one is called Exa. E-X-A, if you're unfamiliar with this company, just put a note for yourself to go to their website later today, or even have your agent to go to the site for you. Exa is similar to Google, but it's built for agents. Okay. It is built for agents. So it means that as agents are searching the web and fetching information, searching the internet and reading web pages that search and fetch, it is built natively for agents to be able to pull the information that humans are looking for.

So Exa works as a great connector and plug-in to turbocharge what you have your agents doing. The next one that I like to use is Apify, A-P-I-F-Y. Again, if you're unfamiliar with this tool, please bookmark it and check it out later. There are a couple of websites that don't have connectors. Okay. One of them being one of my favorite tools, LinkedIn. LinkedIn doesn't share the MCP and API. I'm getting a little technical here, but that's simply how one system talks to another. LinkedIn doesn't readily share that information. So how then do we find publicly available information if our chief of staff agent is helping us prep for a meeting and looking at a company LinkedIn page or the person you're meeting with looking at their public page?

Apify is an agent native tool that fetches specific information from various platforms such as LinkedIn. Now Apify looks at other websites, but I found it to be one of the best if you're prepping and you're using LinkedIn to prepare for a meeting, Apify is one of the best tools for that because they have written a protocol for agents to be able to look at public pages. If it's a private page, it will not read it, and also comments for LinkedIn. So the next tool that I want to share with you that I really like for agents is Firecrawl. You may start noticing a pattern emerge here. I love to give my agents access to agentic native tools.

So Firecrawl, similar to Exa and Apify, it is a system built to help agents to be able to process information on the web. So similar to how I would go to Google and search for information, my agent would first go to Firecrawl and to be able to go through websites and pull information on my behalf. I typically have my chief of staff agent reviewing my calendar and helping me to decide which meetings I need to spend more time prepping for and which meetings I need to have follow-up actions on. If I'm meeting with an organization, have they been in the press recently? Have they shared a press release?

Is there anything they're celebrating? I use my agent's superpower tools, these connectors, to help me prep for that. Someone said someone's been using learning integrations and plugins connected to Claude. Okay, you created dummy emails, cool. So back to the point here with connectors, Firecrawl, Exa, and Apify are three that I typically use. There are over 300 connectors that you can use inside of Gumloop. Take your time and go through those. I'm going to give my agent access to ones I've already given it access to before, such as my Google Drive, Sheets, Gmail, as well as a calendar, okay?

That would help my agent to be able to get work done for me. This is one of those screens inside of Gumloop that you actually have to click the save button to exit out of the screen, okay? So hit save and you'll notice a confirmation that your connectors are there by reviewing the connectors section. I'm going to pause here for a moment for questions. Are there any questions before we advance forward? Connectors is a big piece of this. Any questions? All right, we're doing great. Feel free to put questions in the chat and also upvote them. So following giving connectors beneath that is something called skills. Now, I get just as excited about skills as I do connectors.

How many of you can relate to this? Sometimes you use AI and it gives you a great result. And the next 15 times you're like WTF AI, why won't you do what you did before? Can anyone relate? Give me an emoji in the chat if you can relate to that. So what happens there is that the agents or the AI is trying to figure out, okay, other people can relate. It's not just me. One of the things that we want to do is, a participant, I see you want to know how to connect via an API if your tool doesn't have a connector. We'll hang on to the end. That can be a bit more complicated. That may take about 15 minutes, but I'm happy to help you at the end.

The simplest way to do that is send me a message in Slack or LinkedIn, and I'm going to reach out to the Gumloop team and Slack them and tell them which system you want. If they can do, they'll pull it in for us. That's super easy. They typically ship in the same day, meaning they will service that ticket right away. Let me back up. I did not mention, I don't work for Gumloop. I should have said this at the top of the hour. I don't work for Gumloop. I found this tool over the summer through Lenny's podcast, and I started using it and have since made myself an ambassador for them. I've created a relationship with Gumloop. I am not an employee.

From my understanding, a participant, from other people in WAAI, we cannot connect this agent to WAAI's Slack because the Slack that we use for WAAI is a free account. I don't think we can do that just yet, but I can confirm for you. Back to the screen. Skills are actually workflows. It tells our agent what to do step by step. We can create skills manually, meaning we can tell as if we were writing SOP for a new hire, we could write that for the agent, we could also use something like a site, what I'm showing on my screen, skills.sh, or we can have the agent create a skill for us.

Let me pause for a quick moment and give a word of caution when it comes to downloading skills off the internet. Prompts can be a few sentences or a few paragraphs. Skills can be pages. I am very hesitant to copy someone's skill offline if I don't have access to read it before I download it. I'm super cautious about this. People can write any set of instructions in a skill and you can inadvertently give away access to data that you don't feel comfortable with. The site that I'm sharing with you, skills.sh, this came from a friend. X is a policymaker in California, former U.S. Special Forces executive.

They're a techie and so X shared this with me as a tool to find skills if you don't know exactly what the skill should be. The way that I use this site, skills.sh, I kind of think of it like I'm going to a restaurant and I'm looking for reviews. I may not look at the reviews if there are only five reviews. I may not look at the restaurant if it has 3.6 stars. However, if I see that a restaurant has 500 reviews and 4.9 stars, I then dig in deeper and read a couple of the reviews before I decide I'm going to book a reservation. I use a similar approach when I'm at skills.sh.

If I notice that a skill has been installed 3.7 million times, I may poke around to understand versus a skill that's not been downloaded at all, I won't even poke around and look at that. I just wanted to show you this as something to have in your back pocket not to use today. A participant, I see your question and I'm going to come to that in just a minute. That is a great question and I'll address that in a moment. The best way to create skills inside of Gumloop is to come back to the dictate feature. What I typically do is I start off by saying I want to create two skills for the COS agent.

Ask me three or four questions to customize my agent's skills. It's just that simple. You can take a screenshot of what I've typed into my agent here to create skills. Once we do this, y'all, we are about 60% done. It's incredible how user-friendly this platform is and we're going to test our agent in just a moment. Type into your screen, I want to create two skills for the COS agent. Ask me three or four questions to customize my agent's skills. I am actually going to give my agent access to my calendar. You'll notice I didn't connect it when I was in the connector section, but I was able to one-click log in here and I'm not going to connect mine to Slack now, so I'm also going to say I'll pause on connecting Slack.

In my initial prompt, I said I'm going to use Slack, but now I'm pausing. I can now tell the agent I'll pause on that so it won't continue to prompt me to ask for that. What I would like you to do now is type into the screen what you see on mine, minus the pause on connecting Slack, and hit the up arrow. What you'll see next is that your agent will begin to think and it should come back and ask you a few questions. So it thinks for a few moments, while it's thinking, I want to call attention to two things. As it's reasoning, you'll see exactly what it's doing. You can double-click into those later into the reasoning and review that.

That's one of the reasons why I like building in this platform. You can build agents inside of ChatGPT and Claude. They don't always allow you to click into the reasoning. It may show you the reasoning, but I can actually read the source that it pulled from. So once your agent gives you a few questions to answer, go ahead and answer those questions. I'll give you a moment to go through and read those and answer. And then we'll move forward. All right. Let's see here. Yes, the recording will be shared afterwards as well as a presentation deck that you can use as a companion to the recording.

Great. All right. I'm going to take myself off a video just for a moment because I see that my face is freezing and you're getting all of Jacqueline Twillie's crazy-screened frozen faces. All right.

a participant, I'll answer your question in just a moment. Great question. So what you'll notice is my agent is continuing to think. It's created two skills for me, an end of day sweep as well as a relationship tracker. And the next thing that I'm going to do is I'm going to tell my agent to update agent instructions. So if you're paying attention to my screen, under the Kimi K3 model that I selected, there's no instructions for the agent yet. So just with three words, update agent instructions, it's going to start thinking again.

While it's thinking, I'd like to demo for you that you can then come into, click into agent skills and it shows you step-by-step what the skills are. One of the things that I like to do with my agents is I give them guardrails. I tell them what they can do and what they can't do. And this is just as important of telling your agent what you want it to achieve. A participant, I see you don't feel comfortable moving fast for this build. Totally understand, continue to follow along. We only have an hour. So unfortunately I can't move too slow on this, but I want you to feel comfortable.

That's exactly the right way to delete an agent. So you found a feature that I'm going to demonstrate in just a moment. I'm just asking because, sorry, I totally understand that you have to move fast. But as you see, there was an error. So maybe if you show it, if it works, maybe there's something with my accounts, but sure, please continue. Yes. I want to show in just a moment how we can delete access to connectors as well as delete the agent. So you're a little bit ahead, but I'm going to demonstrate that for us as well. All right. So if we're on the screen now, you'll see I've shown you where that we can double click into skills to read that. I told my agent a few moments ago to update the agent instructions.

You'll now see, if you click into that box, my agent has operating rules. It has core jobs and it has a rhythm. So all of this is editable. The easiest way to edit would be inside of the chat bot. So inside of the chat feature, that's the easiest way to edit both skills and instructions. And I just want to reiterate here, put a reminder on your calendar at a frequent cadence to go in and read agent instructions if you give it access to auto update. So one of the cool things about this platform is compared to other AI systems, you get a lot of control over what it has access to.

So I'm going to click under the connectors section, Gmail. I'm clicking on Gmail under connectors. What happens is it opens up a new screen. This allows me to have extra clarity and control over what my agent does. Some people may have heard of the term harness for AI agents. This is a non-technical way to manage what the agent does. So under tool management, I'm going to select read only tools. There are six options underneath there. I can then restrict or give permission to my agent on what I wanted to read only. Beneath that section, there are write/delete tools.

And in that section, I can then give permission or restrict access to sending emails, updating, creating drafts, forwarding emails, et cetera. One of the reasons why I like this feature, many of the other tools don't yet provide this level of detail when you're giving your AI system access to a tool on what you can restrict and give permissions for. So that's where we would find that. I'm now going to go back to the connectors section. Let's say I want to switch my email account on Gmail to my business account. On the connectors section under Gmail, on the far right side, there is a circle with a check mark and next to it three dots.

If I click on the three dots, it allows me to switch an account, but it also gives me access to easily remove access. So there, I can easily remove access from Gmail. So just like that, my Gmail access has been removed. I'm going to re-add it for the purpose of demo, but I wanted to demonstrate how you remove access. It's also important for me to note here, if you change your password in any system that you've given your Gumloop access to, your Gumloop will no longer be able to access that system. So you would have to come back into Gumloop and reauthorize that permission for it to continue to work. So whenever you change your password in a system connected to Gumloop, you have to come back to Gumloop and reauthorize permissions.

It won't automatically update once those passwords are changed. So I'm going to open up for questions in just a moment and run through some of the questions in the chat. The next thing to do with our agent is run a test. So another three word prompt. Run a test is where we're allowing our agent to then think and provide us with our first result to which we should give feedback. So if your agent comes back and it gives you a result that you love, tell it why you love the results. If your agent comes back and it gives you a result and you have mixed feelings about it, then you'll tell it, "Hey, I like this, but I don't like that," and tell it why.

If you absolutely don't like the results, also tell the agent why. So don't say try again, actually say, "Hey, you surfaced this and this isn't beneficial and this is why." Again, if we think about how we onboard a really competent new hire, the level of instruction we would provide, we'd want to do something similar for our agents. All right, so your agent should think for a few minutes. A participant, you had a question about deleting an agent. So if you come to the top of the screen, I'm hovering over the top sensor section of my Gumloop. There is a chief of staff dashboard. Next to it is the G icon and the number of credits I've used setting up this agent.

Now 681 credits is not a lot for setting up agents. Gumloop uses credits instead of tokens and inside of the Gumloop university section, it explains that it calculates credits similar to how OpenAI and ChatGPT and Gemini calculate tokens. I can provide that resource for you if you want that later, but it's very similar. They just use credits instead of tokens. Now to delete this, next to the box that has an up arrow and a half open box, there are three dots. If you were to click on the three dots, you would be able to delete the agent. If you see an error, what I recommend that you do is refresh your screen.

Sometimes when you're deleting an agent, it actually already went through and deleted. So the first thing I would do is refresh. In most cases, that actually will confirm that the agent has been deleted. If you refresh and it doesn't confirm that it's deleted, simply go back to that box. One of the reasons why, another reason why I appreciate the Gumloop team, they're not trying to trap you in or trick you. They make it really easy to remove access and really easy to delete agents. So just as simple as it was to create, they make it equally simple to delete. And I think that's why they've established a lot of trust with enterprise level clients, because they're not pulling any shady stuff or making you jump through hoops.

Trust is really important for the founders, Max and Rahul. And so they make it really transparent and easy for us to be able to navigate those aspects. So your agent may still be thinking, but if you've gotten a test result, please jump into the chat and tell me what your test result is. And I'm going to run through and start answering questions. Let's see here. There is a lot of freedom for AI agent with personal use, but I'm often restricted when it comes to enterprise data at my company. How do your company's policy security requirements and policy rules shape how you connect at work? What does that compare to your personal use or your own business? Just curious how others are implementing agents or evolving company rules.

So this is really interesting. I mentioned to you at the top of the hour that since the beginning of the summer, I've been doing this for free one to three times a week. And I just put it on my LinkedIn and Threads personal accounts. The amount of C-suite from Fortune 50 companies, including legal team members from Fortune 50 who have joined these workshops has absolutely blown my mind. I typically hand that over to the Gumloop team because I am not an employee of Gumloop, but I have shared the privacy policy and the security link. One of the things that they put together is their trust center screen.

If you're looking at my screen now, you have this link in the chat, you can screenshot Gumloop's security email account, but they also provide a compliance review. So SOC 2, GDPR and HIPAA, there is compliance there. And typically when enterprise level legal teams join my trainings, this is one of the resources that they use as they begin their vetting process to give their organization access to an orchestrator like Gumloop. Gumloop is used by Shopify, Instacart and many other large enterprises. So their team can walk you through that. So if you have enterprise level questions, feel free to message me and I'll connect you with their go-to-market team.

Let's see, someone else says another enterprise question. Yes, I can connect you to the team for anything on that area. Okay. A participant wants to know how does Gumloop compare to Copilot? I am just learning Copilot, but from what a friend has shown, a friend who, one of my friends who's an expert in Copilot, she also uses Gumloop, Gumloop gives you a lot more control and permissions to navigate systems and to orchestrate your tools. So I think it is similar, but I am not an expert in Copilot to be able to compare and contrast each feature, but I can connect you with someone who is an expert in both and that might give you a great viewpoint a participant from someone who lives in both systems.

All right, a participant, did I answer your question? Are you still trying to delete your agent? No, no, I just put in the chat, so I refresh the page and it's fine, but I have another question if I don't want to have another subscriptions because, you know, I tend to have a lot. How do you just, it is just, I found it under my settings, you can just delete the account and that will cancel the subscription after 14 days, so my credit card doesn't get hit. Yes, yes. So at the top of the hour, I mentioned I'd recommend that you set a calendar reminder 12 days from now so that you have a few days to like actually see the system, but cancel it two days before the charge would hit, so you have no issues with that.

Yeah, I heard what I heard that you say that by just by deleting my account will just cancel because sometimes you can just sign out of, you know, you don't have to cancel, I mean to delete the account, but that's the way to not get charged. Well, there's two ways to do that. So one, I would actually cancel the subscription and then I would delete the account. I know some systems allow you just to delete, and that takes care, but I would actually, I would do, I would follow their protocol just so the system doesn't miss that you're actually canceling. Right, it's because I didn't see cancel subscriptions, so that's okay, I mean, I'll figure it out.

And I'll send you step by step on how to cancel because I'm definitely not trying to lock you in here. So I'll walk you through step by step how to do that. Any other questions? Alright, so in terms of test results, give me a thumbs up if you got a test result from your agent. Can I show you what my dashboard looks like? And do I have it create spreadsheet? Yes, I have my agents create spreadsheets too. So let's see, create a spreadsheet with, let's see, because I have people that I'm meeting with this week and I don't want to create a spreadsheet with everyone's name.

Let's see with demo names for a test with fake emails and phone numbers. Alright, so I'm just going to create a spreadsheet here just for the purpose of demo. As I mentioned, I run 32 agents in this system. So I have a lot of information. So whenever I'm sharing my own dashboard or my own systems, I just run a demo so that I'm not sharing too much personal information from the people that I'm working with. So I won't actually jump in the hood and show you my full dashboard. I mean, I just want to be transparent. Why?

Because I don't want to share personal information because I actively use this account. I cannot join as one meeting came over with a record. Yes, you will get a recording as well as access to instructions. So in the recording, we'll send you out step by step on how to build this if you're watching the replay. But if you're watching the replay, of course you can reach out to me inside of the way Slack as well as LinkedIn. I'm happy to answer questions for you. Oh, you can only see the trust center. I'm sorry. Let's see. Let me come back here to share my screen and we're wrapping up here. So I'm happy to do an extended leader basically at time.

So I just want to make sure that we're, you know, we're respecting Jacqueline's time because she's given her time to be here today. I do see some, some questions coming in, but feel free to, you know, Jacqueline is graciously offered to answer them on the WAAI Slack and of course recordings will be available on the program's channel. Give us a few days, but then they will be up and available. And you can also subscribe to our YouTube to get recordings once they come out. I just want to thank everyone for coming today and especially thank Jacqueline for giving of her time. This is such great information and we do have a lot of great upcoming programs, which I'm going to share very quickly so that we can see.

Can everybody see our upcoming programs in case you're not a speed reader, these, this information is always available on the program's channel in, in the WAAI Slack channel. So please do go on if you're ever curious about what's coming up next, you can go on Luma as well, make sure you're subscribed to that calendar so that you can see what's coming up next. And so that you can make sure that you register and save your spot for the great stuff that's coming up. When this ends, you're going to get a survey, please do fill it out, take a few moments and let us know because this is essential information to let us know how we're doing and what we can do better. And I just want to thank everyone today for joining and have a great rest of your week.

Thank you. [end of transcript]