0:00

Welcome to the Prompt Engineering Podcast,

0:02

where we teach you the art of writing effective

0:04

prompts for AI systems like ChatGPT,

0:08

mid Journey Dolly, and More.

0:10

Each week we explore prompting techniques,

0:13

interviews with experts and newbies,

0:15

and tips on selling your prompts. Here's

0:18

your host, Greg Schwartz.

0:19

Welcome to the Prompt Engineering Podcast.

0:21

I'm your host, Greg Schwartz. So

0:25

we have a guest with us today. Go

0:27

ahead and introduce yourself.

0:29

So it's funny, my dad

0:31

was an electrical engineer, so

0:33

he always pushed me for those things. Ended up doing

0:35

some calculus in high school. Then I'm like, dude, I'm

0:37

out. I don't get the jokes anymore. I

0:39

don't belong here. But then I ended

0:41

up going into graphic design, which is for me,

0:44

the perfect combination of like technical

0:46

and creative. So I was able to consolidate

0:49

the two ideas and mourn on the creative.

0:52

And what's interesting about the I got exposed

0:54

to ChatGPTI Alex Hermo

0:57

video, he did a video and

0:59

he's you gotta watch this. And then I watched

1:02

it and it was like, oh my God,

1:04

I gotta get on the, on this. So I

1:06

got it and I started using it. And

1:09

then for me, it's for me, I think

1:12

your experience matters because then you

1:14

know, you have an under. Certain

1:17

questions you're gonna ask, right?

1:19

And so I'm also part of this coaching group,

1:21

and part of the thing that we do is we

1:23

ask better questions.

1:25

Yeah,

1:26

questions lead you to better answers. But

1:29

if you don't know what question to ask,

1:32

you won't get the right answer. So it's a very interesting

1:35

because you don't know what you don't know. I

1:37

have no idea about coding. So

1:39

like, when I see some of these, prompts and things like

1:41

that on the internet, on YouTube, I'm like I wouldn't even

1:43

know where to start to ask. I don't even

1:45

have a language to ask

1:48

those questions, so I don't go there. So

1:50

I ask other things that I'm interested about.

1:53

I ask those kind of questions. And

1:55

because I'm I'm a curious type, like

1:57

I've worked on, like if

1:59

my car doesn't work, I want to get in there and see why

2:01

it doesn't work, and then, That's what

2:03

I do.

2:03

One of the things that I was curious about is what,

2:06

since you were saying you, you aren't an engineer

2:08

and that totally makes sense. A lot of people in this

2:10

aren't. What techniques have you

2:12

been using?

2:14

For instance for my business, so I do graphic

2:16

design and I do websites. So real

2:18

basic emails like I write

2:20

an email out and I say, Hey, can you make it a little bit better? Boom

2:23

that's great. So that helps me with that

2:25

and organizing my thoughts. For instance,

2:27

if I have a convers, discovery conversations

2:29

with a new client, I write

2:32

all this stuff down, all these ideas,

2:34

and I basically just punch 'em in the ChatGPT

2:36

and so it goes through and organizes this,

2:38

and then I see it. I'm like, okay, can can

2:40

you make it a little bit less formal? Can you talk about

2:42

this? Can you add this? Can you take, so I'm.

2:45

It's been amazing, like for proposals,

2:48

like I hate proposals, but this is I

2:50

can get all my random thoughts and these, organizes

2:52

it all.

2:53

That's awesome. So it sounds like mostly zero

2:56

shot prompting, probably not much in

2:58

the way of role-playing is that.

3:00

Okay. So I did some role playing in the beginning. I

3:02

was like I did some sales training. I'm like, okay,

3:04

pretend you're helping me with sales and

3:06

let's like have, let's have a conversation. You're the

3:08

client. I'm the, the salesperson

3:10

and that I'm asking you question. To

3:13

get to, to know you a little bit better. So

3:15

I, I did that a few times, so that

3:17

was interesting. And then it gets a little wobbly

3:20

and I'm like, I gotta get it back. So I try

3:22

to get the program back on track.

3:25

Yes. Yes, totally.

3:27

There's two views of people out there.

3:30

Friends of mine, they're like, oh my gosh, the apocalypse

3:32

is gonna kill us all. That's the

3:34

one side and the other side. They're like, okay

3:37

let's like figure out how this thing works, so

3:39

I always tell people like, listen, apocalypse aside,

3:41

I'm gonna figure out how this thing works. It's interesting

3:44

to me.

3:44

Yeah. What originally got you into

3:47

prompting and ChatGPT?

3:50

Alex Hermo, he's all over the internet.

3:52

He puts a video out his,

3:54

he says, you gotta check this out. So

3:56

I was like glued to his explaining what

3:58

ChatGPT was about. I'm like, that's like

4:02

for me, if I hear something that like.

4:05

Takes a hold of me in such a powerful

4:08

way. I'm like I'm on it. I'll stop when I'm, I'll start,

4:10

just, I'm logging in, I'm starting,

4:12

I'll get on it, and so that's why I got it like

4:15

this last weekend, this interesting. I was like

4:17

I got it in my noggin I wanna do NFTs.

4:20

So I use ChatGPT is okay, if I'm

4:23

a beginner wanting to do NFTs, what

4:25

is the process? What is the steps I need

4:27

take? And so boom. You gotta get

4:29

a crypto wallet, you gotta get this, you gotta get this,

4:32

and then give some suggestions. And then I'm like, okay, what

4:34

is this? What is this? And I ask a question.

4:36

And then within an hour I was on I had a

4:38

NFTs up, like I

4:40

got the crypto, the Coinbase,

4:42

I got this thing, I got the thing on the browser,

4:45

and I got the NFT three, like NFT

4:47

websites, and I did some art. I

4:50

loaded it up there and that was like really amazing.

4:52

Like it helped me like do this within an hour

4:54

or so.

4:56

That's awesome.

4:57

And that's amazing. I

4:59

could do stuff I could learn about things

5:01

and I, I asked you questions like what is like

5:03

the other last night I saw I'm

5:05

a weirder I saw we're watching movies

5:07

and these guys clink their glasses. I'm

5:09

like, okay, where did that? So

5:12

I asked, I said, where did Clinking come from and what is that

5:14

all about? So he told me the history, and then it

5:16

was all about the clink and the G, they would

5:18

spill into each other's glasses. So they would

5:20

know that it wasn't poisoned.

5:23

Is that actually true? I don't know.

5:25

That's one idea that it came up with and it had some

5:27

other ideas. But that was like, that's an interesting,

5:29

so I do things like that. Something gets in

5:31

my head, I'm like, what is that?

5:32

It sounds like you're doing less of I

5:35

am crafting a prompt for

5:37

a specific thing and I've been working on it for three

5:39

weeks or whatever, and

5:41

more of just the, I have a question

5:43

or I have a problem, or I have a project like the

5:45

NFTs and I wanna learn about it.

5:48

Is that right?

5:48

Yeah. And also, the weight training, it's yeah, I wanna

5:50

create myself a six week program so

5:53

I don't have to write it all down and do the

5:55

math and percentages and Excel

5:57

files, and I just wanna do this. And that's what I'm

5:59

working on. If I can create myself a six week program

6:02

strength training, which is heavy on Monday's,

6:05

light on Wednesdays, and medium on Fridays

6:08

with these different exercises with warmups

6:10

and all those things then it's like I can punch

6:12

it in. I could create programs for people

6:15

based on these prompts.

6:17

So I'm putting it on screen now. I'm

6:19

curious to hear, how did

6:22

you iterate on this? I'm noticing

6:24

some pretty specific, you're putting

6:26

in the starting weights, you're, giving

6:28

it specific instructions about like when

6:30

to do different things.

6:32

So basically this program is a linear

6:34

progression. So with squats, you start with squats.

6:37

On Monday, 200 pounds say okay,

6:39

on Wednesday you do like light, 80%

6:41

less, but, and on Friday

6:43

you do medium. And

6:46

so actually I'm 51 years old.

6:48

I'm intermediate. If you're a young at

6:50

20 something, you're basically gonna be adding five pounds

6:53

every time you do squats three times a week, you

6:55

go to a 200, 205, 200.

6:57

Monday, 215. 200 2200

7:00

like you'll add and you'll get huge, like

7:02

I started doing this stuff at 45

7:04

pounds, now I'm up to 210

7:06

pounds squatting.

7:08

Nice.

7:09

Okay. So you get a certain level, like intermediate

7:11

like you plateau a little bit

7:14

and you this age, so you have to mix it up

7:16

a little. So I'm trying to do this sort

7:18

of, heavy light, medium on

7:20

Monday, Wednesday, Fridays. And that's what

7:22

I was trying to. There is an actual

7:24

program that does it. I just don't wanna have to do the math.

7:27

I just wanna be able to say, okay, start

7:29

here in six weeks, I'm gonna end up here.

7:31

So the numbers keep adding up over

7:33

the six week period.

7:35

That totally makes sense. Okay. Since we've got people

7:37

both watching and listening, The watchers are

7:39

able to see this, but the listeners can't, obviously.

7:42

So let me just read this prompt off real quick.

7:45

As your strength training coach, I will design

7:47

a three by five program based on the Starting

7:50

Strength Program by Mark Reito

7:52

that suits your requirements. You

7:54

are a 51 year old intermediate lifter

7:56

looking for a heavy, light, medium program,

7:59

and you want to squat 200 pounds on

8:01

Mondays, Wednesdays, and Friday. Deadlifts

8:04

will be on Wednesdays only on

8:06

odd weeks. You want to bench press on

8:08

Mondays and Fridays and overhead press on Wednesdays,

8:11

on even weeks. Oops,

8:14

on even weeks. You want to do overhead

8:16

press on Mondays and Fridays and bench on

8:18

Wednesdays. Your starting weights are 200 pounds

8:20

for squat, 115 pounds for overhead,

8:23

160 pounds for bench press and

8:25

296 pounds for deadlift. Each

8:28

lift will have a warmup set starting

8:30

at 45 pounds with three reps, three,

8:33

three sets of five reps, one

8:35

set of three reps, and one set of two reps.

8:37

After the warmup sets, you'll do three sets

8:40

of five working sets for each lift. This

8:42

program is specifically designed for an

8:44

older intermediate lifter like you. So

8:47

I'm curious, just have if, go ahead and read

8:49

out the some of the output.

8:51

It did do good. It

8:53

got a little confused, honestly. For

8:55

instance, I wanted to do Bench on Monday

8:59

and Fridays and overhead on Wednesdays. And then

9:01

the next week you switch it. So you do overhead

9:03

on Mondays and Fridays and bench on Wednesdays.

9:05

But it wasn't quite understanding that, so

9:08

I, I ran into a wall, not a wall, but

9:10

I didn't quite complete my task

9:12

or get it to do it the right way.

9:15

I was gonna do week one and week two and give it an example

9:17

and then say, okay, now do this for the

9:19

next six weeks. Figure out a different way to

9:21

to tackle the problem. Because it was, I

9:24

think I confused it.

9:25

That makes sense. And just

9:28

for, the audience, that's an example

9:30

of giving a shot prompt. So

9:32

we've talked a little bit about shot prompting on

9:35

previous episodes, shot prompting is either,

9:38

I'm not including any examples of the output,

9:40

which is called Zero Shot. I'm

9:42

including exactly one. His

9:44

example of here is what Monday

9:46

of week one would look like and it's, this

9:48

lift, this number of weights, et cetera.

9:51

And then few shot prompting

9:53

is, let me obviously give a bunch of those.

9:55

I think sometimes it needs it.

9:57

Yeah, it definitely does. And

9:59

did it help to add that in this case?

10:02

Wait I haven't, I also, I was actually trying

10:04

to do it while I was waiting for you to get on call, so

10:06

I haven't finished. I think it's gonna, I

10:09

think it's gonna work. I just haven't completed the

10:11

task, which I'm gonna do this week.

10:14

Cool. So just for the listeners

10:16

here's an example of the output. Here's

10:18

your revised program with squats

10:21

performed on all three days and deadlifts only

10:23

on Wednesdays. Obviously it's still getting confused,

10:25

but week one, Monday,

10:27

squat three by five at 200 pounds.

10:29

Warmup sets 45 pounds, five

10:32

reps 95 pounds, five reps, 135

10:34

pounds, five reps. Wednesday,

10:37

here's your, so I'm not gonna read this whole

10:39

Yeah. No

10:39

get the idea.

10:40

It's like it's geeking out for the

10:42

weight training crowd. If you look at this, your eyes are

10:44

gonna roll back in your head and you're gonna, just,

10:46

you might as well tell me, explain code

10:48

to me. I'll have the same reaction.

10:51

So actually one trick that I have seen

10:54

that I actually haven't personally used, it just

10:56

hasn't been relevant yet, but is

10:58

asking ChatGPT to format

11:00

the output as a table. And that might be useful

11:02

for you in this case, just so that you're

11:04

not, here's a paragraph and have to break up

11:06

each of the things by day and,

11:09

it was doing that originally. I asked it to do

11:11

the columns

11:12

Nice.

11:13

and then at some point it's no, I'm not gonna do columns anymore.

11:15

I think I'm gonna do your list. You know

11:17

it, it's it

11:18

as always, yeah. Once things

11:20

roll out of context, it'll start

11:22

forgetting and you have to go in

11:25

a table please.

11:26

that's right. That's right. But the tables are very useful

11:29

and it's, I was like, even the first

11:31

reiteration, when it spit

11:33

out the numbers, it was like, it wasn't very

11:35

impressive to see that rollout and because,

11:39

I know what the numbers should look like,

11:41

so I could see that it was adding the five pounds

11:43

and the 10 pounds to the deadlift. So

11:45

it was doing its job. It was doing it, and

11:47

it, it wasn't perfect, but the fact

11:49

that it was cranking out these numbers,

11:51

I was like, dude, this is cool. Getting

11:54

pumped. I'm getting pumped.

11:56

Yeah. Getting pumped for weightlifting. Yeah, absolutely.

11:59

And I've only been doing weightlifting for three or four

12:01

years. So I was like never a gym person.

12:03

Gym bro. I'm not a gym bro. But

12:06

it's as you get older, you need to get fricking,

12:08

your muscle mass and your bone doesn and it goes down.

12:10

So you need I don't wanna fall down, break my hip.

12:12

Good call. Yeah,

12:14

good call. In

12:16

the vast experience

12:18

and I'm joking when I say that cuz earlier

12:21

before we got on the recording, we were chatting

12:23

about how almost everyone I've talked

12:25

to is oh yeah, I have two months of experience doing this."

12:27

"I have six months of experience." But in,

12:29

in your vast experience, what

12:31

would you say are the key factors in

12:34

making a prompt successful?

12:36

As a veteran of three months I would say,

12:39

that's funny right there. I would say

12:42

you have to remain curious. You have to be like,

12:44

this is just, it's a tool. And

12:46

in order for you to succeed, it's like

12:49

you have to stay curious. You have to keep asking

12:51

questions you have to ask better questions.

12:53

You have to it, it helps if you know what you, you

12:55

want the answer to be. So it's basically just

12:57

massaging it talking to it and asking

13:00

better, different questions to, to get what. It,

13:03

it takes time, but it's if

13:05

you're not fascinated by this process, you're

13:07

not gonna like doing this. But if you're

13:09

curious and you're like, oh, interested,

13:11

if you're geeking out on this kind of stuff you'll be there

13:14

for hours, doing this. And it's, I

13:16

think, stay curious my friend.

13:19

I can definitely agree with that. I've had

13:21

many a late night of "Wait, it's what time?"

13:25

Yeah. We're not alone on that one. I've heard that from a

13:27

lot of people. What

13:29

would you recommend to people who are starting

13:32

with prompt engineering? As so many

13:34

of the listeners are where they're just like, ChatGPT

13:37

sounds cool, but I don't really know how to use

13:39

it. Like, where would they start?

13:41

I would start with what is it that you do

13:43

and you're interested in trying to basically

13:47

like figure out a way to use

13:49

ChatGPT to help you do what you do.

13:51

Whether it's your hobby, your business, your, your

13:53

interest, anything. It could be anything just.

13:56

Figure out a way or come up with a way to

13:58

help you do a, an aspect

14:00

of it. Even if it's

14:02

like, Hey, this thing helps me write better emails.

14:05

Boom. I don't have to I hate typing.

14:08

I do this because this is how I feel when I'm typing.

14:11

I don't have to, like I'm

14:13

some very T-rex like

14:15

yeah. That's how it feels. Like I'm the clucking. I'm,

14:17

English is second language. I don't know how to spell things

14:19

over and over. Though just

14:22

certain words, I just don't get it. For years

14:24

now. So start where you, what you know, and then

14:26

build from there. And then as you go keep watching

14:28

videos, keep watching podcasts. Be interested

14:30

and find a niche or something interested

14:32

and start, it's like a a thread on

14:34

a sweater. Like you find something, you're like, oh, what's

14:36

that? And you start pulling that thing and all of

14:39

a sudden you discovered a whole nother, old

14:41

take your. That you can't even

14:43

imagine,

14:44

Nice. So your prompting,

14:46

are you doing it in a

14:48

prompt engine? Are you doing it in the playground?

14:50

Are you doing it in just the ChatGPT

14:53

sort of web app?

14:55

ChatGPT if there's something else I should be in

14:57

I'm, send me link. I'll sign up. You

15:00

said there's like a playground? Is that something, that's

15:02

something I'm.

15:03

So playground is a more

15:06

technical version of

15:08

the ChatGPT interface. It's

15:10

got a bunch of interesting controls

15:12

that you can use around what are called temperature

15:15

and top p, which basically

15:17

are ways of saying how random

15:20

do you want the answers to be? If you're doing math,

15:23

you want basically no randomness cuz you want

15:25

the actual answer. But if. I

15:28

don't know, brainstorming what to write

15:30

for marketing copy, for example, you can turn the randomness

15:33

up because you don't want tell

15:35

me about problems that people have who are,

15:38

guitar players in it to be like selling

15:40

tickets yeah, no, duh, tell me something

15:42

else. So you turn up the randomness and

15:44

it'll be like, people, I don't know what

15:46

it's gonna say, but, people stealing their sheet music

15:48

or something like that. It's higher randomness

15:51

leads to more creativity. But

15:53

also more noise. So yeah,

15:55

it's a double-edged sword.

15:57

I think giving ChatGPT context

15:59

is important, I think that's, that's helped a lot

16:01

If you give it more information on background about

16:03

where you going. I actually saw this video yesterday. It

16:05

was very interesting. It says, then you ask

16:07

it if you have any questions before you start.

16:09

And then it will ask you if he

16:12

has clarifying questions and then

16:14

you can run the program. So that's very interesting.

16:16

So I'm watching ChatGPT videos

16:19

while I'm waiting, resting between

16:21

sets. So I'm like, this, I got chalk

16:23

everywhere, so yeah. Context is

16:25

important.

16:25

The whole, Here's what you, I want you to do,

16:27

but first ask me questions to improve

16:30

the output. That nuance

16:32

is really useful. I've seen that used on a few

16:34

prompts.

16:36

It's fascinating, man. I mean, the fact that

16:38

there's already jobs that people do. There's a part

16:40

of me that's. Fomo, there's

16:42

a FOMO aspect to it. I don't wanna miss out

16:44

like it, I'm interested in what this will

16:46

do to my business, am I gonna be out

16:49

of a job? Rather than be afraid

16:51

of it, I'm embracing

16:53

it because I'm like, I want to know what's happening

16:56

and where it's going, and I want to

16:58

be, on top. I just, I guess I wanna be on top

17:00

of kind of tracking it so

17:03

I can make moves if I need. If

17:06

I need to pivot in my business, if I need to get out

17:08

of the business or whatever it is I'm supposed

17:10

to be doing, at least I have a my eye on something,

17:12

then I can make a more informed decision.

17:15

That totally makes sense.

17:17

It feels like I sh I should be doing that.

17:19

Like I, I don't know how people are, like,

17:22

I'm afraid of people who are like, oh, it's evil,

17:25

and it's, it's gonna take over. I'm like, yeah, those

17:27

are. Concerns

17:29

that are valid, but don't just

17:32

ignore it.

17:34

have you sold any prompts

17:36

or bought any prompts?

17:37

No I just got here.

17:39

No worries. I ask because I actually

17:42

created a sales analysis for people on

17:44

Prompt Base. So prompt base for context

17:47

is a website where you can sell

17:49

and purchase prompts. And

17:52

I'm selling one about Writing

17:54

a good cover letter where you paste

17:57

in your resume, you paste in the job description,

17:59

and then it outputs, here's a cover letter

18:01

for that job position using your

18:03

resume details. Anyway when

18:06

I was thinking about how to build that, or

18:08

really when I was thinking about what do I want to build,

18:11

I went and did this analysis

18:13

where I took all of the data on prompt

18:15

base and basically said, okay, where

18:17

are the niches that are earning?

18:20

A fair amount of money, but don't have a ton

18:22

of competition because it turns

18:24

out some of them, like marketing as

18:26

a whole, it has I don't even remember how many hundred

18:29

prompts

18:30

Oh, gotcha. So marketing

18:32

a, it's kind A saturated,

18:34

exactly, at least on prompt base, maybe

18:37

in other areas it's not a saturated, but

18:39

I was specifically thinking, I wanna sell a prompt

18:41

on prop base, so where

18:43

should I focus my energy? So

18:46

anyway I did this analysis. It is available

18:48

on Gumroad at https://gregschwartz.gumroad.com. People listening,

18:50

I will put a link to it, but

18:53

yeah, that's why I've been curious about, if you're selling

18:55

or if you're buying any, how did you think

18:57

about that? So it's okay that you aren't doing

18:59

that but feel free, it's actually pretty awesome.

19:02

if I could figure out these uh, training course, training

19:04

prompt. There's like ton

19:07

of different like programs and that

19:09

actually exist on an Excel

19:12

file

19:13

That would actually be, yeah I bet there would be people

19:15

who are like, I don't wanna figure out all the

19:17

but the Excel files do the math. So you punch

19:19

in the numbers and it does it all for you. But

19:22

maybe it doesn't matter. I don't know. Who

19:25

knows?

19:25

It's definitely worth trying. I know there are

19:27

fitness prompts. I don't remember

19:29

if there are any specifically around

19:32

weightlifting. Most of them are around, give

19:34

me six exercises to do at home,

19:36

or, more nutrition

19:38

focused stuff. Here's what's in my fridge, tell

19:40

me what I can cook, kind of things. Is there anything in

19:43

the news that has you particularly excited

19:46

that you've heard about recently?

19:47

I guess the increase in quality of ChatGPT

19:50

tutorials and, insights and things. So that,

19:52

that, that is exciting. It's also terrifying

19:56

to see where this could go.

19:58

It's like there's a terrifying and a exciting

20:01

aspect, both at the same time. This is terrifying,

20:03

but this is super exciting. Oh my God, we're all

20:05

gonna die. But this is really amazing. So SkyNet's

20:09

coming up online tomorrow. But wait, I gotta

20:11

sell a couple prompts before then. I

20:13

dunno.

20:15

Yes, I know what you mean.

20:16

Just the

20:19

more we focus on bad things the more I.

20:22

Anxious and anxiety and anxious and

20:25

it shuts me down. So it's a trick

20:27

to be able to like, keep your peace

20:30

do what you are supposed to be doing. What is it I'm supposed

20:32

to do today? What is that excites me?

20:35

What is it that brings me joy?

20:37

Do that right? And then, We're

20:41

all gonna die. Okay, that's a given.

20:43

Fine. We're all gonna die. We don't know when, but we're

20:45

all gonna die. But so remain in peace

20:47

and do the things you're meant to do in life.

20:49

Yeah, and the FOMO thing is real,

20:52

not falling for FOMO

20:53

yeah, just, find the joy, find

20:55

the thing that brings you joy, and go do that.

20:58

And then forget about the clickbait stuff.

21:00

If I can make you laugh like I made you, And

21:03

to me that's regardless of whether

21:05

I sell a prompt or whatever, if I have a I got

21:07

to meet you, Greg. And that's

21:09

the, actually, that was the most important part of

21:11

this whole thing. If me getting on ChatGPTto

21:14

get to talk to you was what it was all about

21:16

then dude I'm good. You know what I mean? That's,

21:19

that's how I feel about this podcast as a

21:21

whole is, I want to educate people about

21:24

here's how to write good prompts and here's some really specific

21:26

techniques, but also just connecting

21:28

with people is so much fun. So

21:31

yeah. Totally with you.

21:32

This's not. This is a good conversation. Hopefully

21:34

it was valuable to you and it brings value to

21:36

your listeners. I, I love podcasts, so I'm

21:39

like, I constantly listen to podcasts, so being

21:41

a part of a conversation to me is super awesome.

21:44

Do you have any links, social

21:46

media, anything like that you wanna

21:48

Oh, I have my website I could actually

21:50

I could put in, it's EnvisionaryDesign.com

21:53

and that's my business. And from there's Instagram

21:56

and Facebook and things like that. I don't tweet.

21:58

I haven't figured that one out. Maybe ChatGPT

22:00

can help me tweet better.

22:02

I think it can. I think it can. Awesome.

22:06

All right, I will put your website in the description

22:08

and thank you so

22:09

Greg.

22:11

Thanks for coming to the Prompt Engineering podcast,

22:13

the podcast dedicated to helping you learn

22:16

how to be a better prompt engineer, how

22:18

to sell your prompts if that's something you're into,

22:20

or just use them in your day job. See

22:22

you soon.