WEBVTT - How I AI: Agents Explained in 10 Minutes (No Jargon, No Hype)

0:00:00.880 --> 0:00:03.080
<v Speaker 1>Hey, just quickly from me before we get into it.

0:00:03.200 --> 0:00:06.360
<v Speaker 1>If you're using AI but still doing too much manually,

0:00:06.519 --> 0:00:08.879
<v Speaker 1>Neo and I have built something for you. It's called

0:00:08.920 --> 0:00:11.320
<v Speaker 1>the AI Agent boot Camp, and it's a four week

0:00:11.440 --> 0:00:14.840
<v Speaker 1>virtual program and you'll walk away with agents already built

0:00:14.880 --> 0:00:18.160
<v Speaker 1>and working. Spot's limited and all the details are in

0:00:18.200 --> 0:00:22.480
<v Speaker 1>the show notes. You've probably heard the word agents thrown

0:00:22.560 --> 0:00:25.880
<v Speaker 1>around a lot lately, and if you've been nodding along

0:00:26.239 --> 0:00:30.400
<v Speaker 1>while quietly having no idea what it actually means, you're

0:00:30.480 --> 0:00:34.479
<v Speaker 1>not alone. The AI companies have done a genuinely terrible

0:00:34.560 --> 0:00:39.600
<v Speaker 1>job of naming things agents, agentic, GPTs, projects, gems. It's

0:00:39.680 --> 0:00:43.000
<v Speaker 1>a lot of words for concepts that, once you understand them,

0:00:43.240 --> 0:00:47.159
<v Speaker 1>are actually pretty straightforward. In this episode, Neo and I

0:00:47.240 --> 0:00:50.240
<v Speaker 1>cut through all of that. We cover what an agent

0:00:50.320 --> 0:00:54.440
<v Speaker 1>actually is and why it's different from agentic AI, which

0:00:54.480 --> 0:00:57.600
<v Speaker 1>is a distinction that's important to know, and we get

0:00:57.600 --> 0:01:01.000
<v Speaker 1>into how both individuals and teams can start aren't using them,

0:01:01.200 --> 0:01:04.040
<v Speaker 1>from automating meeting minutes to building a knowledge agent that

0:01:04.080 --> 0:01:07.480
<v Speaker 1>fields your team's repetitive questions so that you don't have to.

0:01:08.319 --> 0:01:10.920
<v Speaker 1>By the end, you'll know exactly what an agent is

0:01:11.160 --> 0:01:13.560
<v Speaker 1>and have a clear sense of where to start building

0:01:13.600 --> 0:01:22.959
<v Speaker 1>one for yourself. Welcome to how IAI with me Doctor

0:01:23.000 --> 0:01:27.720
<v Speaker 1>Amantha Imba and Neo Applin, head of Inventium AI. Each

0:01:27.760 --> 0:01:31.040
<v Speaker 1>episode we share one practical way to use AI better

0:01:31.120 --> 0:01:34.920
<v Speaker 1>at work and in life. No fluff, no tech jargon,

0:01:35.240 --> 0:01:39.160
<v Speaker 1>just things you can use straight away. Neo, I want

0:01:39.160 --> 0:01:45.360
<v Speaker 1>to talk about agents today, and I'm sure there's lots

0:01:45.400 --> 0:01:47.160
<v Speaker 1>of people that know what an agent is, but then

0:01:47.240 --> 0:01:49.919
<v Speaker 1>there are an awful lot of people that are probably

0:01:50.000 --> 0:01:52.520
<v Speaker 1>quite confused as to what is an agent? What is

0:01:52.560 --> 0:01:55.440
<v Speaker 1>a gentic? Is there a difference? Is that the same thing?

0:01:55.800 --> 0:01:58.360
<v Speaker 1>How does it relate to my work life? So let's

0:01:58.400 --> 0:02:02.720
<v Speaker 1>start off with some definitions. What is an agent?

0:02:04.480 --> 0:02:06.360
<v Speaker 2>The first thing I'm going to say is the way

0:02:06.400 --> 0:02:09.840
<v Speaker 2>that these companies have named these things is terrible. There

0:02:09.840 --> 0:02:12.400
<v Speaker 2>are so many confusing names out there as projects as gpds,

0:02:12.400 --> 0:02:15.160
<v Speaker 2>there's agents, there's agentic and all those kind of things.

0:02:15.240 --> 0:02:18.720
<v Speaker 2>So here's how I like to look at it. An

0:02:18.760 --> 0:02:21.960
<v Speaker 2>agent is generally a thing that we will use. I'm

0:02:22.000 --> 0:02:24.800
<v Speaker 2>talking now about an agent, a GPT or a project

0:02:25.960 --> 0:02:28.960
<v Speaker 2>to help you to do something as a reusable expert.

0:02:29.040 --> 0:02:31.080
<v Speaker 2>In other words, help me to do the thing in

0:02:31.120 --> 0:02:34.920
<v Speaker 2>the same way every single time. And what it's not

0:02:35.200 --> 0:02:37.360
<v Speaker 2>is to do the entire thing. So it's not like

0:02:37.400 --> 0:02:41.000
<v Speaker 2>an entire workflow. It's just to do one thing. So

0:02:41.040 --> 0:02:43.840
<v Speaker 2>you might have an agent that maybe checks your work.

0:02:44.040 --> 0:02:46.519
<v Speaker 2>It might be an agent that potentially you drop in

0:02:46.560 --> 0:02:48.760
<v Speaker 2>the meeting transcript and will give you the meeting action

0:02:48.880 --> 0:02:50.440
<v Speaker 2>items and notes and all those kind of things the

0:02:50.440 --> 0:02:52.800
<v Speaker 2>way that you want your action items and notes to

0:02:52.800 --> 0:02:53.200
<v Speaker 2>be done.

0:02:53.440 --> 0:03:00.600
<v Speaker 1>So it's an agent, okay, And is that different from AGENTICAI.

0:03:00.160 --> 0:03:03.959
<v Speaker 2>To go to. Yes, So agentic is where it does

0:03:04.120 --> 0:03:07.600
<v Speaker 2>the things not in the same way every single time.

0:03:07.880 --> 0:03:10.280
<v Speaker 2>It's kind of got a goal and it may have

0:03:10.360 --> 0:03:12.480
<v Speaker 2>some tools and we'll figure out how it's going to

0:03:12.600 --> 0:03:15.120
<v Speaker 2>reach that goal using the tools, and you leave it

0:03:15.160 --> 0:03:18.160
<v Speaker 2>to that thing to decide what to do. So, for example,

0:03:18.160 --> 0:03:21.480
<v Speaker 2>you might have an urgentic thing which has an email

0:03:21.560 --> 0:03:23.240
<v Speaker 2>tool so it can email in and out and read

0:03:23.240 --> 0:03:25.360
<v Speaker 2>emails and things like that. It might have file access

0:03:25.400 --> 0:03:27.520
<v Speaker 2>so it can then access your files and so you

0:03:27.520 --> 0:03:29.079
<v Speaker 2>can then give it a goal. Hey, you do this thing,

0:03:29.280 --> 0:03:31.400
<v Speaker 2>and it might read emails, it might read files, it

0:03:31.440 --> 0:03:34.440
<v Speaker 2>might figure outs and things, might get attach a file

0:03:34.480 --> 0:03:36.040
<v Speaker 2>to an email and send it off, so it's like

0:03:36.360 --> 0:03:39.320
<v Speaker 2>it will decide what to do. Now, with agents, it's

0:03:39.360 --> 0:03:41.600
<v Speaker 2>more about you needed to be bound to do the

0:03:41.640 --> 0:03:44.800
<v Speaker 2>same thing the same way every single time, and it's

0:03:44.880 --> 0:03:48.800
<v Speaker 2>often a smaller thing to achieve in type for example,

0:03:48.880 --> 0:03:51.440
<v Speaker 2>like a Riunning style guide checkup. You know, it just

0:03:51.440 --> 0:03:53.680
<v Speaker 2>does one thing, does it really well? Does it reliably

0:03:53.720 --> 0:03:54.520
<v Speaker 2>every single time?

0:03:55.360 --> 0:03:58.440
<v Speaker 1>And today and for the next few weeks, we are

0:03:58.440 --> 0:04:01.160
<v Speaker 1>going to be focusing on a gents as opposed to

0:04:01.520 --> 0:04:06.680
<v Speaker 1>agentic AI. So when it comes to agents, depending on

0:04:06.960 --> 0:04:10.320
<v Speaker 1>what AI tool you're mostly using, can you took us

0:04:10.320 --> 0:04:12.920
<v Speaker 1>through like what the equivalents are where they we're in

0:04:13.200 --> 0:04:16.880
<v Speaker 1>co pilot versus Claude versus Chattipit versus Gemini.

0:04:17.240 --> 0:04:20.040
<v Speaker 2>Yeah, Before I do that, I just want to quickly

0:04:20.080 --> 0:04:23.080
<v Speaker 2>say you do not need to be a geeky rocket

0:04:23.120 --> 0:04:26.600
<v Speaker 2>scientist AI a kind of person to be able to

0:04:26.600 --> 0:04:32.440
<v Speaker 2>build an agent AGENTIC Yep. Maybe that's something very different workflows, yep. Maybe,

0:04:32.520 --> 0:04:35.600
<v Speaker 2>but this is all reasonably easy for just normal regular

0:04:35.680 --> 0:04:37.840
<v Speaker 2>human beings to be able to do so in the

0:04:37.880 --> 0:04:41.479
<v Speaker 2>different tools they are called in Copilot's called an agent.

0:04:41.560 --> 0:04:43.120
<v Speaker 2>That's one I'm going to start there because they've actually

0:04:43.120 --> 0:04:47.280
<v Speaker 2>called it that. In Chichipitt it's called a GPT. In

0:04:47.600 --> 0:04:53.120
<v Speaker 2>Gemini they're called gems, and in Claude generally one ones

0:04:53.160 --> 0:04:57.400
<v Speaker 2>we're talking about where we're talking about actually projects. Interestingly enough,

0:04:57.440 --> 0:05:00.560
<v Speaker 2>projects or another kind of term. Chechipt also has projects

0:05:00.680 --> 0:05:03.480
<v Speaker 2>as a similar kind of thing, and also there are

0:05:03.520 --> 0:05:06.960
<v Speaker 2>notebooks in a copilot as well. So these are all

0:05:07.160 --> 0:05:11.040
<v Speaker 2>different flavors of agents, and the thing that they've all

0:05:11.080 --> 0:05:13.680
<v Speaker 2>got in common is you can give it an instruction

0:05:13.800 --> 0:05:15.960
<v Speaker 2>in other words, a prompt, a small or a large prompt.

0:05:16.000 --> 0:05:17.840
<v Speaker 2>Here's what I want you to do. And you can

0:05:17.880 --> 0:05:20.640
<v Speaker 2>also give it some files so that it has some knowledge.

0:05:20.680 --> 0:05:22.599
<v Speaker 2>Maybe it's like a here's a template to use, or

0:05:22.640 --> 0:05:24.719
<v Speaker 2>it might be here's some knowledge that I need from you.

0:05:24.760 --> 0:05:28.719
<v Speaker 2>So it's instructions and a file or files, and that's

0:05:28.800 --> 0:05:31.160
<v Speaker 2>all you are giving your agent so it can do

0:05:31.200 --> 0:05:32.760
<v Speaker 2>the thing. So it doesn't have a bunch of tools

0:05:32.800 --> 0:05:35.200
<v Speaker 2>and it can't just go things by itself. And it's

0:05:35.200 --> 0:05:37.640
<v Speaker 2>also very responsive like you. It can only do things

0:05:37.680 --> 0:05:40.440
<v Speaker 2>when someone asks it to do the thing. It doesn't

0:05:40.440 --> 0:05:44.200
<v Speaker 2>do things at three in the morning. But yeah, so agents, GPTs,

0:05:44.440 --> 0:05:47.279
<v Speaker 2>projects and gems are generally the names of these things.

0:05:48.480 --> 0:05:51.600
<v Speaker 1>Let's talk about how individuals can use them, and then

0:05:51.600 --> 0:05:53.799
<v Speaker 1>I want to talk about how teams can use agents.

0:05:53.839 --> 0:05:56.839
<v Speaker 1>And for the purposes of today, we're just going to

0:05:56.960 --> 0:06:00.039
<v Speaker 1>use the term agents. So what are some of the

0:06:00.040 --> 0:06:05.040
<v Speaker 1>way that individuals can get value from creating an agent to.

0:06:05.040 --> 0:06:07.000
<v Speaker 2>Do the things that you do all the time just

0:06:07.000 --> 0:06:09.599
<v Speaker 2>to make your life better. So if you are that

0:06:10.200 --> 0:06:12.760
<v Speaker 2>unfortunately I've done this before, and it sucks to be

0:06:12.800 --> 0:06:14.640
<v Speaker 2>that person who has to get the meeting transcript and

0:06:14.640 --> 0:06:16.520
<v Speaker 2>then turn out the minutes and then check the minutes

0:06:16.560 --> 0:06:18.960
<v Speaker 2>and then email them out to people. If that's something

0:06:19.000 --> 0:06:21.440
<v Speaker 2>you do every single week or you've got multiple meetings,

0:06:21.640 --> 0:06:23.280
<v Speaker 2>it's great to build an agent to be able to

0:06:23.320 --> 0:06:25.279
<v Speaker 2>do that because it then speeds up your time and

0:06:25.279 --> 0:06:27.560
<v Speaker 2>all you got to do is check what it's done.

0:06:27.880 --> 0:06:31.520
<v Speaker 2>So it's like a really helpful small assistant there. You

0:06:31.600 --> 0:06:36.680
<v Speaker 2>might have something like a voice and style checkert if

0:06:36.680 --> 0:06:38.880
<v Speaker 2>you'd like. Maybe that's a company one, but maybe that's

0:06:38.920 --> 0:06:42.320
<v Speaker 2>even just yours, where you might say, hey, i've got

0:06:42.320 --> 0:06:44.480
<v Speaker 2>some text here, I've got to look after the website.

0:06:44.560 --> 0:06:47.279
<v Speaker 2>Maybe you're not a marketing kind of person and you've

0:06:47.279 --> 0:06:48.920
<v Speaker 2>given a couple of dot points and said, now turn

0:06:48.960 --> 0:06:50.920
<v Speaker 2>this into Vivee voice and style and sort of then

0:06:51.040 --> 0:06:54.640
<v Speaker 2>do things like that also good for like bulk emails

0:06:54.839 --> 0:06:56.400
<v Speaker 2>and things like that, just to make sure it's actually

0:06:56.480 --> 0:06:58.760
<v Speaker 2>hitting the right beats. So it's something you just do

0:06:58.880 --> 0:07:02.800
<v Speaker 2>small and often. Other ones might be research assistance. So

0:07:03.120 --> 0:07:05.760
<v Speaker 2>I've got one which is a research prompt builder. I

0:07:05.800 --> 0:07:08.920
<v Speaker 2>think we'll talk about that in another episode. But yeah,

0:07:09.400 --> 0:07:11.640
<v Speaker 2>when I'm doing researching, I do a lot of research

0:07:11.760 --> 0:07:13.160
<v Speaker 2>just to make sure I'm keeping up to date on

0:07:13.240 --> 0:07:15.880
<v Speaker 2>different industries and how AI is going and things like that,

0:07:16.080 --> 0:07:18.560
<v Speaker 2>and so I've got a research prompt buildup that helps

0:07:18.600 --> 0:07:21.080
<v Speaker 2>me to build those things up. So the things that

0:07:21.160 --> 0:07:23.560
<v Speaker 2>you do more than I'm going to say once or

0:07:23.600 --> 0:07:26.960
<v Speaker 2>twice a week every single week, ask yourself, is this

0:07:27.040 --> 0:07:29.400
<v Speaker 2>something that you can put into an agent? Because your

0:07:29.480 --> 0:07:31.480
<v Speaker 2>job is going to be so different to mine and Amantha's,

0:07:31.880 --> 0:07:34.160
<v Speaker 2>you'll figure out what you're going to do more often.

0:07:34.480 --> 0:07:36.200
<v Speaker 2>Maybe that's something you can build into an agent.

0:07:36.720 --> 0:07:41.120
<v Speaker 1>Now let's look at teams. How can teams get benefit

0:07:41.400 --> 0:07:44.360
<v Speaker 1>from agents similar.

0:07:44.120 --> 0:07:46.640
<v Speaker 2>Kind of things in order to make their lives better?

0:07:46.680 --> 0:07:49.560
<v Speaker 2>And so some of them is about the things that

0:07:49.600 --> 0:07:51.720
<v Speaker 2>you do that you have to do all the time

0:07:52.120 --> 0:07:56.760
<v Speaker 2>as a team. So maybe meeting minutes is really handy thing.

0:07:56.800 --> 0:07:59.280
<v Speaker 2>Maybe multiple people are doing meeting minutes. There's that, but

0:07:59.360 --> 0:08:02.239
<v Speaker 2>maybe there's all things like reports. So there's a way

0:08:02.360 --> 0:08:06.640
<v Speaker 2>that we produce those types of reports. And sometimes what

0:08:06.680 --> 0:08:09.560
<v Speaker 2>you want to do is have that same report being

0:08:10.080 --> 0:08:11.760
<v Speaker 2>kind of built the same way, even though it may

0:08:11.760 --> 0:08:13.960
<v Speaker 2>be built by different people. And here's the thing that

0:08:14.240 --> 0:08:15.840
<v Speaker 2>I might have a different style the way I'll do

0:08:15.920 --> 0:08:17.800
<v Speaker 2>my report, you'll do your style to the way you

0:08:17.880 --> 0:08:19.760
<v Speaker 2>do your report. It would be great to have an

0:08:19.800 --> 0:08:22.720
<v Speaker 2>agent that has like the approved way that it grabs

0:08:22.760 --> 0:08:25.360
<v Speaker 2>these data in and then crafts it and then puts

0:08:25.360 --> 0:08:27.480
<v Speaker 2>it into the right headings and things like that. So

0:08:27.520 --> 0:08:30.280
<v Speaker 2>the things that you and your team do often I

0:08:30.320 --> 0:08:32.520
<v Speaker 2>think they're really good things to do to build as

0:08:32.559 --> 0:08:34.920
<v Speaker 2>an agent. The other thing is the things you wish

0:08:35.000 --> 0:08:37.600
<v Speaker 2>you wouldn't do, So how can you offload some of

0:08:37.640 --> 0:08:41.520
<v Speaker 2>those things to AI? So that's another one. And the

0:08:41.559 --> 0:08:45.000
<v Speaker 2>third is the questions that you often have or that

0:08:45.040 --> 0:08:48.600
<v Speaker 2>you get as a team. So if you're in HR,

0:08:48.679 --> 0:08:50.840
<v Speaker 2>I imagine you get an awful lot of questions on

0:08:51.040 --> 0:08:53.480
<v Speaker 2>how do I change my bank details? When's the next

0:08:53.480 --> 0:08:56.920
<v Speaker 2>pay run? Those kind of questions. If you're in like

0:08:56.960 --> 0:08:58.920
<v Speaker 2>a customer service role, you're going to get lots of

0:08:59.000 --> 0:09:01.840
<v Speaker 2>questions on this thing is broken, how do I fix it?

0:09:01.880 --> 0:09:03.520
<v Speaker 2>And then you'll probably know a lot of that in

0:09:03.520 --> 0:09:05.920
<v Speaker 2>your brain. But wouldn't be better if there was an

0:09:06.000 --> 0:09:09.040
<v Speaker 2>agent that did a look up of your company procedures

0:09:09.160 --> 0:09:12.640
<v Speaker 2>or the FAQs and things like that. And of course

0:09:12.760 --> 0:09:15.320
<v Speaker 2>here's that the other thing, which is why are you

0:09:15.559 --> 0:09:18.880
<v Speaker 2>answering this questions? Wouldn't be possible to build an agent

0:09:19.040 --> 0:09:21.120
<v Speaker 2>that answers those questions for those people, so you don't

0:09:21.120 --> 0:09:23.160
<v Speaker 2>even have to get involved. So it's really useful for

0:09:23.200 --> 0:09:26.360
<v Speaker 2>those kind of things. So yeah, if it's often regular

0:09:26.520 --> 0:09:28.839
<v Speaker 2>you do it the same way as a team, or

0:09:28.880 --> 0:09:31.120
<v Speaker 2>people are pestering you and your team, then maybe what

0:09:31.160 --> 0:09:33.160
<v Speaker 2>you can do is create an agent feed at the

0:09:33.240 --> 0:09:35.640
<v Speaker 2>knowledge with the files and things like that so that

0:09:35.640 --> 0:09:37.240
<v Speaker 2>people can self service those answers.

0:09:38.080 --> 0:09:39.800
<v Speaker 1>So what now and I are going to be doing

0:09:40.000 --> 0:09:42.640
<v Speaker 1>over the next few weeks is we are going to

0:09:42.640 --> 0:09:45.640
<v Speaker 1>be digging deeper into agents. So we are going to

0:09:45.640 --> 0:09:49.120
<v Speaker 1>be unpacking things like, if you're a knowledge worker, what

0:09:49.160 --> 0:09:52.440
<v Speaker 1>are the must have agents for all knowledge workers. We

0:09:52.640 --> 0:09:54.680
<v Speaker 1>are going to go into some of the tasks that

0:09:54.720 --> 0:09:58.280
<v Speaker 1>you should never do manually again, we are going to

0:09:58.320 --> 0:10:01.560
<v Speaker 1>go into Neo's favorite AG, the one that he uses

0:10:02.000 --> 0:10:05.160
<v Speaker 1>pretty much every single day. And we're also going to

0:10:05.160 --> 0:10:07.280
<v Speaker 1>get into how can you build a knowledge agent that

0:10:07.520 --> 0:10:11.040
<v Speaker 1>answers every one oh one question like there was just

0:10:11.080 --> 0:10:13.880
<v Speaker 1>talking about, so that you don't have to So if

0:10:13.920 --> 0:10:16.120
<v Speaker 1>you want to go deeper into the world of agents,

0:10:16.240 --> 0:10:19.640
<v Speaker 1>make sure you have hit subscribe or follow, because we

0:10:19.800 --> 0:10:22.760
<v Speaker 1>will be sharing a lot more info about agents over

0:10:22.800 --> 0:10:26.480
<v Speaker 1>the coming weeks. How i AI was hosted by me

0:10:26.640 --> 0:10:29.480
<v Speaker 1>Amantha Imber and Neo Applan. A big thank you to

0:10:29.559 --> 0:10:32.800
<v Speaker 1>Martin Imba who does our sound editing, and Jim Rubio

0:10:32.920 --> 0:10:36.040
<v Speaker 1>for production support, and thank you to John Kilby who

0:10:36.080 --> 0:10:37.400
<v Speaker 1>composed the theme music.