101 AI Apps You Can Vibe Code

Here are 101 AI apps that you can start

building today and you don't even need

to know how to code. Just as a note, I'm

not saying that coding is obsolete. It

is still very much necessary and useful

for more complex and custom apps and

features. But it is very much possible

now to build certain types of

applications without traditional coding

aka bip coding. I have done many myself.

So I'm going to split this video into

seven different categories. Starting off

with database/data management type apps,

hardwarebased apps, dashboards,

chatbot/ass assistant agent type apps,

coaches/arning agent type apps,

multimodality, and finally

automation/mros. Now, before we dive

into all the different app ideas, I do

want to do a quick crash course on my

five-step framework for how to approach

building AI applications all the way

from ideiation to deployment. This will

give you a practical starting point for

how to build these projects and the

tools to use for this when you actually

get really excited to build these

applications. A portion of this video is

sponsored by Bault. The five-step

framework is metaprompt, product

requirements prompt, the PRP,

implementation debugging and

deployment. The first part of the

framework is the meta prompt. And this

is what I like to use in the very

beginning in the ideation phase because

it goes through and asks you questions

like what's the purpose of the app?

Who's the target audience? Who are you

marketing? What are the core features?

And it goes through all of these and

forces you to think about who it is that

your application is actually designed

for. Because after all, if you don't

even know exactly what it is that you're

building, it's kind of unfair to expect

your AI to figure out how to build it,

right? So, I'm actually going to provide

you the meta prompt, which I'll link in

the description that I like to

personally use. Okay. So, after you

think very deeply and answer all of

these questions, the output of this meta

prompt is going to be called a product

requirements prompt, which is the second

part of this framework, a PRP. If you

guys have heard about a product

requirements document before, it's

generally what a lot of product managers

will use to have a very clear

description of what their product is and

all the core features surrounding it. So

that meta prompt will help you generate

this PRP. And what you do is that you

take this PRP and you input it into your

AI assisted coding tool. Just with a

single prompt, you'll probably be able

to get like 80 to 90% of your core set

of features, which is honestly really

really impressive. But to get the rest

of the 10 20% to actually make it into a

full application, we do need to go on to

the next third step which is the step of

incremental implementation. You see, for

the things that it doesn't get quite

right, you do need to incrementally

start asking it to change certain things

like add a little bit more of this or

change a little bit more of this, change

the UI a little bit more. Um, you know,

add something else like maybe add like

an audio functionality. And actually

step three and step four go hand in hand

and that is debugging. as you're working

with the AI and getting it to implement

more things. You're also going to be

coming across errors or things that the

AI is just not doing correctly. This is

really really normal and part of the

process. So when you do encounter these

errors, the first thing that you want to

do is just try to ask AI to fix it

itself. Literally just be like there's

usually a button that you can just click

and be like, you know, try to fix the

error itself. Um or you can do things

like take a screenshot of the error or

just point out that something is not

working like this button is not working.

Then the AI is usually pretty good at

going back and then fixing these errors.

This is also going to be incremental. So

you're going to be repeating this

process of incrementally implementing a

feature and then debugging it,

implementing debugging implementing

debugging until towards the end, you get

to a point in which you're happy with

the features that have been implemented

and it's functioning properly. Then at

that point, you're ready for step five,

which is deployment. Honestly, there are

a lot of things that I can talk about in

this deployment section, including

things like security, risk, version

control, or to be hosting things. But

the good news is that for most of these

VIP coding AI assisted coding tools,

they do have their own deployment

options. Okay, so that is a really

really quick crash course on building an

AI application all the way from ideation

to the deployment stage. There is so

much more that I can talk about and I

actually made like full videos going

through this process which I will link

above over here and also in the

description if you do want to go into

more detail about these step-by-step

processes. But for the rest of this

video, I do want to focus on the AI apps

that we are very excited to start

building. Starting off with database/

data management type apps.

This category of AI applications is all

about leveraging AI's incredible ability

of turning raw or unstructured data into

something that you can search, query,

summarize, and extract value from. The

general workflow for this type of

application usually starts with

ingesting a lot of different types of

information, labeling it, indexing it,

and then organizing it into a nice

little database. then allow users to be

able to ask questions and the AI would

go and filter all the information in the

database to collect the relevant

information maybe do some types of

analysis if it has to and finally output

it in some format maybe like

visualizations an email or a report for

example if you're building something

like a video search database and a user

can come in and ask I want to get the

specific clip of this specific math

question that was being shown by the

professor uh through like catalog of

hours and hours of lectures please

please please somebody built this by the

way the workflow what it would look like

is that you would first give it like a

batch of different lectures all the

different lectures that are available.

The AI would like transcribe that

information, restructure it, index it,

and put it into a database. The user

would be able to ask a question like

find the specific formula from all of

these different videos. The AI would

search through the video content,

extract that clip, and then show it to

the user where they can like, I don't

know, download it or export it,

whatever, right? This kind of database

search application is also really

powerful for other types of modalities

like images, quotes, text, and of

course, just combinations of these

different things as well. You can also

make a cross tool enterprise search

where an AI is going to be connected to

the database sources from like notion

and Slack and Google Drive and a lot of

different things and you can unify all

of that as a search engine. Audio

transcription and search is a really big

one if you have a lot of podcasts and a

lot of meetings that you have there and

you just want to pick out a very

specific section and also semantic

search over internal documents being

able to search very specific things

related to an internal company system.

Sometimes the emphasis of an app can be

on the transformation of that data,

allowing the user to upload a bunch of

PDFs and notes and then AI being able to

organize that information um in terms of

topics or Q&A sections. Data cleaning

assistant. Maybe you have a bunch of

Excel files that have like just columns

and all over the place like everything

is all over the place. There's missing

values. Everything's not transformed

correctly. You can build an AI app

that's able to clean everything up and

just allow the user to redownload the

clean version. Image tagging or

classification. Maybe you have like lots

of different pictures about flowers and

you want to upload that and then allow

the AI to classify it into different

types of flowers. Works for all

different types of modalities. You can

also create an app that annotates all

the files that are being uploaded. Like

if it sees a picture of a butterfly, it

would actually put an annotation and a

tag on it as well as specific keywords

related to that. Say for example, if you

have something like medical imaging,

it's able to annotate what it considers

to be risk or no risk and then pass it

along to the appropriate healthcare

provider. You might want to create an

app that is focused on audio and speech

components so that people are just able

to ask it directly like what were sales

in Asia in Q2 like and the app will be

able to have a conversation with you

about the data. And finally, you can

build an internal app for reports and

reporting. It cannot be underestimated

how much time most companies spend just

on generating reports and passing along

to either internally or to different

stakeholders. You can make an app that

automates this process. Everybody will

thank you because nobody likes making

reports. I think the next category is

hardware related applications.

[Music]

This is a really cool category and I

feel like a lot of people don't think

about it as much like how much

opportunity there is. They're always

thinking more just on the software side.

But AI is very much able to interact

with the physical world. For this type

of application, the workflow usually has

an emphasis of things being real time

and looping. The general workflow is

that you collect some type of real-time

data usually from some type of physical

device like a sensor, a camera, a

wearable. The AI would process and

analyze that information to detect

things like patterns or anomalies or

different insights. It will decide if

something stands out from that

continuous stream of information like

trigger an alert or change something.

And finally, it would summarize and then

report whatever it is that it did. An

example of an AI app in this category

would be a traffic incident detector.

You will have cameras and road sensors

feeding in information, continuous

information to your AI. The AI would be

analyzing that real-time information.

And if it detects something like, oh,

like there is a stalled vehicle where

like a car crash, it would make a

decision. Oh, like is this car crash

really bad? Is it actually congesting

things? And if it does decide that it's

a problem, it would then take action.

Like it would probably report an alert

to an operator and then maybe do things

like adjust the traffic lights to try to

like mitigate the situation. and then it

would finally summarize it saying, "Oh,

there's an accident that occurred on

like road 7 and then give that to

somebody." Speaking of things related to

traffic, you can also make a traffic

flow predictor. So, it's able to take

historical and live data and be able to

forecast different congestions and

reroute drivers. Illegal parking

detections like having cameras that's

able to spark cars that are parked

incorrectly and then I guess like be

able to find them directly or alert

them. crowd management sensors that are

detecting the flow of pedestrian traffic

and be able to flag different

bottlenecks and then like maybe send

people who would go, you know, those

people who kind of like wave their

sticks to redirect people, like get

those people to go if there's a lot of

congestion. air/noise pollution apps for

cities, being able to use sensors and AI

to detect where the hot spots are,

predict spikes are going to be there,

and be able to help like urban planners

or like developers figure out where

certain buildings should be built and

how to like plan the city better. For

more general security and camera related

apps, you can build something like an

object/person recognition application.

So, your AI is able to identify

packages, pets, vehicles, or known

people and be able to flag that. AI

privacy filters like it's able to blur

out certain things that you don't want

to be seen. Sometimes maybe you want to

upload a video of something but there's

like you know identifying information

like people's faces or like license

plates and things like that. So you can

automatically blur that out. Anomaly

detection and camera feeds. The anomaly

would be different depending on the

different types of scenario but you can

get the AI to be able to detect what you

consider to be an anomaly in a lot of

different types of feeds and CCTV

cameras and send it to the popo. There's

a lot of applications related to smart

home devices, too. Like energy anomaly

detection. Maybe you can have smart

plugs that's able to, you know, look at

your fridge and be like, "Oh, there's a

15% more power usage next month." And

then, I don't know, like miticate that

in some fashion. Appliances that would

know when it should start itself. Voice

activated appliances. Appliances that

are able to detect when it should be

started and when it should be stopped.

>> Hello, robot.

>> Yes, I'm here.

>> Turn on the light.

>> Okay. a smart fridge inventory that can

remind you to buy things or just

directly order itself when you know

inventory is low on milk. An AI that's

able to detect when something is about

to fail or some sort of sensor has

failed. Like if you have a glucose

monitor, if something is going wrong

with it, it's able to send an alert.

Speaking about wearables and healthcare

related things, you can have a wearables

aggregator application that's able to

take all your different types of

wearables like your Aurora link, your

Whoop, your Apple Health, whatever.

Integrate all of them together and give

you insights about your daily activity,

how to improve your life. For example,

you should stop eating things late at

night because it increases your stress

response and then you don't sleep very

well. A sleep and diet app that's able

to figure out what is the right plan for

you based upon your current lifestyle.

smart accessibility devices like AI

enabled hearing aids that's able to

increase or decrease certain sounds and

filter it so you can hear better.

>> These systems are a great benefit to the

heart of hearing and rapidly increasing

around the world. Assisted listening

benefits a large and ever growing

section of society. Hearing impairment

affects one in seven of us, a number

that is increasing the population ages.

>> Smart glasses that has cameras and can

like tell you certain things. Kane

sensors so you're able to have

information about your surroundings.

apps that are integrated into your car,

being able to tell you when it is that

you should get things checked up in a

much more direct fashion, not like just

an ominous blinking on your on your car

close fleet monitoring like device

vehicles or drones. Being able to

monitor their different patterns and

then adjust them. The next category is

dashboards.

[Music]

AI enhanced dashboards are all about

having real time or AI enhanced

overviews of specific types of

information. It allows users to be able

to get an overview of certain systems

and then also doubleclick and get more

detailed information if they choose to.

The emphasis is on how the information

is being portrayed for the user. The

general workflow of a dashboard starts

off with gathering different data

sources. And this data source is usually

near real-time data sources. This can be

through internal information that people

are updating. So like CSV files or

databases. It can also include external

information from different places. Then

there's the data processing and analysis

which is the cleaning and restructuring

of that data followed by the most

important part which is the compilation

and visualization. So compile everything

together showing it to the user. You

might also want to help the user out by

highlighting like the most important

things they should be paying attention

to at a glance. And finally

distribution. How it is that you want

the viewer to be accessing that

information. Should it be through an

email? Should it just be like a link

that you send out to people? A daily

report linking back to the dashboard. A

classic example of an AI app in the

dashboard category is a personal finance

dashboard. The app is able to pull the

bank statements, different types of

transactions from a lot of different

places that you're spending your money,

process and analyze that information and

categorizing the different types of

transactions. Then compiling it together

and visualizing it like showing you

different graphs and charts of how

you're spending your money and the

different trends over time. It will

generate insights for you like, ooh,

careful there. You just spent 30% of

your monthly budget on chocolate.

Probably not a good idea. And finally,

distribution. Maybe you want to have

like a WhatsApp message from your

dashboard saying whenever there's like

certain things that are going on, maybe

you want an email summary or even like a

voice briefing about your spending

habits. Another really classic and

useful AI app use case is for building a

dashboard that can monitor certain

trends. For example, if you're a

company, you might be interested in your

competitor's doings, such as what their

marketing stuff is happening, their

sales projections, competitor news,

industry shifts. In my case, I have an

AI news aggregator dashboard that I look

to see what are the happenings in the AI

world and it's very customized towards

things I am specifically interested in.

I of course also track to creators in my

niche as well to see what kinds of

videos and content that they are

producing within a company. Building an

AI powered KPI internal tracker

dashboard is extremely powerful. There

is a saying that what doesn't get

tracked doesn't get done. You can create

a dashboard for customer sentiment

looking at all the reviews from

different users and then compile them

together by product. Anomaly detection

dashboard tracking if there's any types

of weird spending patterns. Website

errors or metric spikes. Cash flow

forecasting dashboard. Taking real-time

information for accounting and

transaction data making sure that we're

not overspending or under spending.

Pricing intelligence dashboard. Scraping

competitor prices and helping you adjust

things that are either underpriced or

overpriced. Sales pipeline dashboards

with CRM integrations. Subscription

churns where you're inputting user

behavior data. Dashboards for resource

allocation. Is the staffing actually

correct? Where should we be allocating

different projects? Are we really

spending money on the right things? Bug/

error prioritization. Which bugs are

actually the biggest priority to be

addressed? Cyber threat. Maybe there's

threatening social media information

about your company where there's like

attacks that are happening. You want to

be monitoring those. And finally, a

little passion project that I'm working

on, which is an investments dashboard.

So, it tracks different assets like real

estate or commodities like gold for

example, as well as like stocks and

bonds. and then it helps me like figure

out what I should be investing in and

what my trading strategies are. Now, if

you're feeling inspired to start

building some of these applications, let

me introduce you to Bault. It's the

first unified platform that takes you

from idea to live product in minutes. No

code required. Whether you're technical

or not, you can go from scratch to

launch without worrying about setup or

config files. Everything is included

right inside Bault Cloud. Hosting,

backend, database, file storage, O

domains, and even SEO. You can also plug

in integrations like Superbase for data

and stripe for payments and deploy

instantly without ever leaving Vault.

It's built to scale too with

enterprisegrade performance with

partners like Netlefi. You can go from

your first user to millions without ever

changing platforms. Here are two real

examples built using Bolt. Ask in Bio

where you can ask creators questions

based on their publicly available data

and another app that lets you create

personalized books and audiobooks for

your child. All no code, all running

completely on Bolt. Bolt Cloud is

currently in private beta and pro users

get full access during the beta for

free. You should try building your own

application based on one of these

suggestions right now at bolt.new. It's

free and instant. The link is in the

description. Thank you so much Bald for

sponsoring this portion of the video.

Now back to the video. Next category is

chatbot/ass assistance.

[Music]

A lot of apps in this category will have

a gent components to it. Not going to go

into too much details about building

agents right now, but I do have a video

which I will link up here that goes into

a lot of detail about building agents

specifically, but agents are just like a

subcategory of AI apps in general. So,

all of this is still applicable for

chatbots and AI assistants. Um, a really

big emphasis here is on the ability to

communicate with the user and then also

being able to perform actions related to

their function. The general workflow of

this is that your user will start off

with some sort of input, some sort of

request. The AI will figure out what the

intent of the user is, go and retrieve

some information from a database and/or

take some type of action and then go

back to the user and give it information

uh that the user was requesting or

informing the user that it has went

ahead and performed some type of action.

Then finally, logging into behavior

somewhere so you know what your chatbot

assistant agent has been doing. For

example, if you have an accounting

assistant, the user can say something

like, "Generate an invoice for Lonely

Octopus for $5,000." The accounting

assistant would go and classify this

request as an invoice creation, pulls

the client information for Lonely

Octopus, and generate the invoice using

some sort of accounting software through

their API and then respond to the user

saying that the invoice is generated and

maybe ask the user, do you want me to go

ahead and send this as an email to

Lonely Octopus and finally logging that

behavior in the internal system so it's

trackable. Um, and then if it's sending

out an email, also saving the email

receipt. You can imagine that you have a

legal assistant that could do something

similar by generating contracts as well

as explaining contracts, explaining

clauses, highlighting different risks.

An internal policy assistant that's able

to answer questions concerning human

resources or compliance questions based

upon internal documents. A compliance

assistant that's able to monitor local

laws and alerts you when your business

might have to adapt. An invoice

assistant that can specifically extract

information from invoices and receipts

that you might be collecting for

different places. And then you can, this

is literally like the bane of my

existence. So currently in the process

of building an AI assistant that can do

this. So it's able to take you can take

pictures of these receipts and then it

would file them automatically for you.

Tax filing assistant. Taxes also the

bane of my existence. So it can file

taxes for you. For small businesses,

having a CFO agent could be really

helpful because people usually don't

like to think so much about that. Um but

it's also very important to do things

like maintaining and monitoring cash

flows and forecasting runway and then

giving warnings before payroll. like I

have had a lot of experience in messing

up these areas. Medical insurance

assistant uh especially in the US

insurance very confusing having a

medical insurance agent that's able to

go through that process with you and

file claims for you. Benefits navigator

if you're working at a company often

times you have a lot of benefits that

you are not aware of. Medication guide

assistant if you have certain

prescriptions you can get an assistant

to help you explain those prescriptions

and fill those prescriptions as well.

Health record organizer. You know, if

you've ever had the experience of going

to different hospitals and they just ask

you like the same questions over and

over again, you can have an assistant

that's able to compile all your health

records as a unified health record

assistant provider. Customer service

agent that's able to handle frequently

asked questions and escalate complex

task when it's required. A meeting

follow-up assistant setting up meetings,

just personal assistants. IT help desk

assistant with your IT problems.

Appointment booking assistants. These

are agents that can actually call up

different places like restaurants like

make reservations or like different

appointments with medical appointments

and then be able to book certain things

for you. Personal shopping assistant,

shopping for food, shopping for clothes,

shopping for home stuff. Subscription

manager assistant. Oh my god, I need

this. I have so many different

subscriptions and it is very difficult

to manage them and I also don't want to

pay another subscription to manage my

subscriptions. You can just build an

agent for that. Warranty and returns

assistant. How many times have I just

given up and like not returned something

because the process of doing that was

too complex? Utility negotiator.

Sometimes you can negotiate a lot of

things like your phone bill, your

internet bill, your water bill. There's

a lot of things you can actually

negotiate and over time it really adds

up. The next category is personalized

coaches.

[Music]

So this is also a category that has a

lot of agentic components to it but it's

different from the general assistance

like chatbot assistance because there's

a bigger emphasis on the learning aspect

and and feedback from the user. The

general workflow for a coach AI

application so your user will be

interacting whether like through text,

voice, video, action or exercise. The AI

would evaluate it like it would score

the whatever it is that the user has

inputed analyze the user's performance

then provide feedback and suggestions

for how to improve. It's also important

for the coach to be able to provide

encouraging reinforcement and then also

provide guidance for next steps. And

finally, having some sort of progress

tracking, be able to track the

progression of the user's learning

through time. An example of this would

be like a relationships coach. The input

could be a user being like, I got angry

at my partner and I was, you know, not

being nice. How do I improve? The AI app

will take this information, ask the user

questions like, how did they handle this

conflict? What exactly was being spoken?

What exactly happened? what is it that

they spoke about? Evaluate it, then

provide feedback for how they can

improve, like suggesting better phrasing

and being more empathetic. The app will

also help reinforce this by suggesting

things like roleplaying so that next

time they're able to apply these

techniques, and progress tracking,

seeing if the user has become better in

their relationship over time. Some other

really cool apps that you can build in

this category would be like a drawing

coach, being able to upload sketches

where just sketching in real time and

then having commentary about how to

improve. I actually built this one. any

type of like sports or physical activity

coach like tennis or golf or just like

working out being able to upload like

video of your golf swing for example and

then the coach would be able to analyze

your golf swing and then give you like

suggestions for how to improve it.

Public speaking coach recording a talk

or saying what your talk is and then

getting real-time feedback and

commentary for how to improve. Language

learning coach, this one is amazing like

having real-time conversations in a

specific language and then getting

feedback for it. I have also built this

one. It's been really great. a career

coach that's able to give you mock

interview practices and just like any

type of like niche specific coach like

pottery or like music or like things

like this like if you actually hired an

expert it would be very expensive to do

so. So if you just like build an

application to do it for the majority of

people they can accomplish like 80 or

90% of their learning through the AI app

before having to actually go to a human

and pay a lot of money to improve. Next

up is multimodality apps.

[Music]

So we have already seen a lot of like

multimodality like text, video, audio

and images like a lot of transformations

already. But this category of apps I did

want to split out by itself because it

is such a strong suit of AI.

Multimodality apps have a big emphasis

on generating and remixing different

types of content. The general workflow

is that a user will have some sort of

input like an idea, a topic or a prompt.

The AI app would generate a draft of the

content and the user would work with the

AI app in order to add different

enhancements and different

personalizations before finally

publishing it and distributing it

through maybe like social media or like

a newsletter or as an email. An example

of an application like this would be

like a slide deck generator. The user

will have some type of input like make a

deck about the history of video games.

The app would generate the slide deck

the core information surrounding that.

Then the user will work with the app in

order to enhance it, personalize it,

change and tweak different parts of it

before finally the user is happy with it

and they might export it as a PowerPoint

slide, download it as PDF or send it

across as an email. This is specifically

for creating slide decks, but you can

also have a very similar workflow for

creating things like newsletters, video

content, podcast content, music, images

for Instagram, social media posts of any

type really. You can also create apps

that analyze this type of content like

analyzing images and photo trends,

analyzing videos, combining these things

together, like being able to input video

and then having the app comment on the

video and give commentary like for a

soccer game for example, giving

commentary about what's happening in the

game.

>> Okay, I see you're talking about a

specific Chrome tab. Is there anything

you would like to know about it?

>> What is happening in this clip right

now? Okay, in the video the players are

moving around the field and a player in

black is on the ground having just been

tackled. It looks like the player in red

and white has gained possession of the

ball. Also, the score is now Arsenal

Zero Western

>> Converter applications like video to

text, text to image, image to video,

etc., etc. An interactive storytelling

app that's able to generate entire

stories and have like accompanying text

and images and videos and audio and

everything. Just playing with these

different types of modalities and

converting them and combine them

together, you can have so many really

cool creative ideas and different types

of niches as well. Now, time for the

final category, which is automations and

macros.

[Music]

This is also a category that is so

diverse and I think people don't think

about them enough as well. If you think

about it, there are like a lot of very

repetitive things that we do on a

day-to-day basis, both in our personal

lives and at work as well. A lot of

these things are all opportunities for

us to automate and use AI to do it. The

general workflow for this type of AI app

is to have some sort of trigger that

might be like an event or like an input

where something basically just happens.

Then the AI would take that trigger and

maybe like extract that information or

take it as a cue in order to perform

some type of task. After it runs the

specific task, then it will log it and

then maybe like communicate it in some

fashion to the user. Although for this

type of app, the goal is to make it as

autonomous as possible. So have it run

and do things without direct input from

the user. Roughly speaking, there's also

two different categories of this type of

application. The first type is workflows

that are automated on the cloud. For

example, you can have a customer

feedback classifier. The trigger here

would be just having a new customer that

arrives through like Zenesk or some type

of CRM. The AI app would automatically

take that review, extract information

about that review, and about the user as

well. Take that and label it maybe in

terms of urgency, like is it really

urgent? Is it something that's very

complex? Is it something that doesn't

require any action? Then if it's

something that maybe is very urgent,

requires action, it could do something

like go and submit a Jira ticket for

customer service to go and fix something

based upon the review. And finally, it

would have a log of this and then would

also report it. The second type of

application in this category is usually

referred to as macros. And these are

applications that are local to your

computer. For example, you can have a

file organizer. The trigger or input

would just be like if you download

certain files on your laptop. I don't

know about you guys, but mine is like a

complete absolute mess. Downloads are

everywhere. That would be the trigger

for your AI app to take that file, uh,

put that into the right folder, and

maybe even rename that file so it can be

easily be found. Like say if you

downloaded an invoice, it would rename

that invoice and put that into the

invoice folder for your company. Then it

would do something like have a record of

that and have a report of that. Some

other types of AI automations that you

can build on the cloud would include

something like an automated email repli

invoice processing meetinguler sales

prospect enricher that's able to take a

sales prospect and add additional

information like their LinkedIn

recruiting pipeline automator that's

able to scan and go through uh different

potential candidates code review

assistant a task generator that's able

to convert chat requests into like a

Jira or a Trello ticket proposal

generator that's able to autodraft

proposals. Compliance checker that's

checking for specific types of

compliance issues and then flagging the

things that are problematic. A CRM

notetaker that's able to log summaries

and then input that into your CRM. Also

would be like meeting notetakers as

well. A cross tool syncing agent that's

able to sync a lot of different types of

applications together so you don't need

to manually do it yourself. For

automation apps that are local, you can

have a PDF/doc summarizer that you're

able to rightclick onto things and it

would instantly summarize things for

you. a clipboard assistant, so you're

able to copy and paste things and it

would expand the text for you. A voice

command agent, so you're able to have

the ability of communicating with your

computer directly through voice, like

telling it to run certain scripts. A

spreadsheet macro agent that's able to

input the common formulas that you

usually use on your spreadsheet.

Screenshot analyzer. If you take a lot

of screenshots, it might automatically

be able to just take the screenshot and

annotate it and then put it in the right

place. Local search assistant, maybe you

have a lot of files and documents and

things like that and you want to search

across your local computer. So, you can

create an agent that's able to do that

for you. A photo tagger. Have a lot of

photos, want to attack the photos.

Presentation helper. Maybe you're

creating a presentation. You wanted to

stay on your computer for security

reasons. A security assistant that's

able to scan your local files before you

upload it onto the internet to make sure

you're not revealing any of your

secrets. So, there you have it. That was

a lot. There are so many different types

of AI applications that you can start

building. I really hope that by watching

this video, you have some inspiration

now and there's certain things that you

haven't thought about and that you're

really excited to start building. Let me

know in the comments what it is that you

want to build out and I will see you

guys in the next video or live stream.

[Music]