00:00:00
Forbes has just listed AI Cloud engineer
00:00:02
as one of the top 10 hottest AI jobs for
00:00:05
2025 and in between the numbers there's
00:00:08
something hidden in front of us that
00:00:10
nobody is talking about right now
00:00:12
there's a massive shift happening in
00:00:14
tech companies are desperately racing to
00:00:17
build AI systems but they're hitting a
00:00:19
major roadblock they can't find
00:00:21
Engineers who understand both Ai and
00:00:24
Cloud infrastructure but how desperate
00:00:26
are they the numbers are staggering
00:00:28
there's currently 5,000 open positions
00:00:31
for AI Cloud Engineers yet only 890
00:00:34
monthly searches for this role for every
00:00:36
person searching for an AI Cloud
00:00:38
engineering job there are five positions
00:00:40
waiting to be filled but here is where
00:00:43
it gets really interesting while
00:00:45
everyone is fighting over the standard
00:00:47
Ai and machine learning jobs where there
00:00:49
are four people competing for every
00:00:51
single position AI Cloud engineering
00:00:53
roles are sitting empty so you know what
00:00:55
does this mean this is the biggest
00:00:57
opportunity in Tech right now the compet
00:00:59
comption for AI Cloud engineering roles
00:01:02
is incredibly low while the rewards are
00:01:04
very lucrative with average salaries
00:01:06
topping over
00:01:07
$120,000 a year but it won't last
00:01:10
forever as more people discover this
00:01:12
hidden gold mine the 5:1 ratio will
00:01:14
disappear quickly so if you want to
00:01:16
seize this golden opportunity and become
00:01:19
an AI Cloud engineer where would you
00:01:21
even start what does an AI Cloud
00:01:23
engineer actually do and most
00:01:25
importantly how do you position yourself
00:01:28
to grab one of these rols before before
00:01:30
everyone else catches on hi I'm slan
00:01:33
I've spent over a decade in Tech and
00:01:34
today I help companies scale and secure
00:01:37
the cloud infrastructure on AWS and if
00:01:39
there's one thing that's taken me years
00:01:41
to learn it's this most people are even
00:01:43
stuck in certifications collecting
00:01:45
mindset or they're in a tutorial hell
00:01:48
never taking any decisive action and
00:01:50
that's what most of my students tell me
00:01:52
before they join my Academy and here is
00:01:54
the advice that I give to them to truly
00:01:56
stand out in today's job market you need
00:01:59
to understand what businesses actually
00:02:01
need and then position yourself as the
00:02:04
person who can deliver it the
00:02:05
opportunity right now is clear companies
00:02:08
are sitting on mountains of data but
00:02:10
there is a problem they lack Engineers
00:02:12
who can turn all of this potential into
00:02:14
real business value and this is why AI
00:02:16
Cloud Engineers are so valuable because
00:02:18
they can build AI solutions that save
00:02:20
hundreds of hours drive millions in
00:02:22
revenue and create highly personalized
00:02:25
experiences that keep their customers
00:02:27
coming back I'm going to show you
00:02:29
exactly how to become one of these
00:02:31
highly paid Engineers using my proven
00:02:33
full-face blueprint you'll discover the
00:02:35
exact technical skills that companies
00:02:37
are desperately hiring for I'll show you
00:02:39
how to build your first AI Cloud project
00:02:42
even if you have zero experience you'll
00:02:44
learn which tools and Frameworks will
00:02:46
put you ahead of 99% of our applicants
00:02:49
and most importantly you'll see how to
00:02:51
position yourself to land these six
00:02:53
figure roles so by the end of this video
00:02:55
you won't just know what an AI Cloud
00:02:56
engineer does you'll have a clear road
00:02:59
map to becoming one of the most in
00:03:01
demand Professionals in Tech today by
00:03:03
the way you can join Over 11,000
00:03:05
Engineers accelerating their Cloud
00:03:07
journey and grab my free Cloud
00:03:08
engineering startup pack Linked In the
00:03:10
description all right so most people
00:03:12
fail to become engineers because they
00:03:14
start in the wrong place they jump
00:03:16
straight into AI Frameworks and complex
00:03:18
cloud services missing the foundation
00:03:20
that makes everything else possible so
00:03:23
in order to build AI Cloud projects you
00:03:25
need to start at it and Cloud
00:03:27
fundamentals specifically Linux the
00:03:29
backbone of cloud computing nearly every
00:03:32
AI system you build will run on Linux
00:03:34
servers so you'll learn to use the
00:03:35
command line and learn how to write
00:03:37
simple scripts to automate repetitive
00:03:39
tasks this naturally will lead into
00:03:41
networking where you discover how these
00:03:43
Linux servers communicate with each
00:03:45
other using IP addresses and how to keep
00:03:48
them secure with firewalls and vpns and
00:03:51
as you're working with these systems
00:03:53
you'll need somewhere to store data and
00:03:55
that's where databases come in you'll
00:03:57
learn both SQL for structured data think
00:04:00
of customer information or inventory and
00:04:02
nosql databases for more flexible
00:04:05
storage needs and finally you'll tie
00:04:08
this all together with virtualization
00:04:10
which lets you create multiple virtual
00:04:11
servers from a single physical machine
00:04:14
the technology that makes cloud
00:04:15
computing possible in the first place
00:04:17
and with these it fundamentals under
00:04:19
your belt we have to choose a cloud
00:04:21
platform and I recommend AWS because
00:04:24
it's the market leader and offers the
00:04:25
most job opportunities they also provide
00:04:28
an extensive free tier perfect for
00:04:30
Learning and experimenting without
00:04:31
having to worry about any of the costs
00:04:34
now in AWS you need to understand five
00:04:36
essential services that work together
00:04:38
and don't worry we're going to implement
00:04:39
AI in just a moment but for now covering
00:04:42
ec2 which gives you the virtual service
00:04:44
where you can run your applications
00:04:46
these applications often need storage
00:04:49
which is where S3 comes in think of it
00:04:51
as an unlimited cloud storage for any
00:04:53
files and folders or images and videos
00:04:56
to keep everything secure you need
00:04:57
identity and access management which
00:04:59
lets you control who can access what and
00:05:01
you know we mentioned SQL and nosql
00:05:04
databases earlier in AWS you'll need to
00:05:06
learn the equivalence of RDS and Dynamo
00:05:09
DB and finally VPC brings this all
00:05:12
together by providing you your own
00:05:14
private Network in the cloud where you
00:05:16
can organize your resources exactly how
00:05:18
you want now I know some of you might be
00:05:20
worried about coding but you shouldn't
00:05:22
be yes you do need to know how to write
00:05:24
code but also you need to know what the
00:05:27
lines of code that you write and review
00:05:29
in codebases even mean so you can't skip
00:05:32
this step you have to learn it now this
00:05:34
will take time and some experience but
00:05:35
this is a journey rather than a
00:05:37
destination now as an AI Cloud engineer
00:05:39
you need to be comfortable with writing
00:05:40
in Python and terapon these are your
00:05:42
essential tools for Automation and
00:05:44
infrastructure management now while
00:05:46
you're learning these Services you
00:05:47
should also get your AWS certifications
00:05:50
especially if you're coming from a
00:05:51
non-technical background you want to
00:05:53
start with a cloud practitioner and then
00:05:55
go into the AI practitioner then you
00:05:57
want to move onto the solution architect
00:05:58
SOA before going to AWS security
00:06:01
specialty this is a very nice and clean
00:06:03
road map now I do also recommend you
00:06:05
looking at the machine learning
00:06:06
associate certification as a bonus but
00:06:08
just remember these certificates should
00:06:10
be seen as complimentary to your
00:06:12
learning and not your main focus now I'm
00:06:14
personally going through these
00:06:15
certification myself so I only recommend
00:06:17
to you guys what I do myself now before
00:06:19
we get to building AI Cloud projects I
00:06:21
need to give you some context on AI
00:06:24
infrastructure what it is and how it's
00:06:26
different to regular Cloud
00:06:27
infrastructure now regular Cloud
00:06:29
infrastructure ructure is what powers
00:06:30
most of the internet today when you use
00:06:33
Netflix check your email or browse
00:06:35
Instagram you're using regular Cloud
00:06:37
infrastructure it runs on standard
00:06:39
computers CPUs organized in data centers
00:06:42
handling things like storing your photos
00:06:45
processing payments or running websites
00:06:47
but AI infrastructure is completely
00:06:49
different while regular Cloud handles
00:06:51
everyday task AI systems need much more
00:06:54
power because they are processing
00:06:55
massive amounts of data and doing all
00:06:58
these complex math calcul ations at once
00:07:00
think of analyzing millions of images or
00:07:02
understanding human language tasks that
00:07:04
need serious computing power and this is
00:07:07
why AI systems need specialized
00:07:09
computers with powerful gpus now AWS
00:07:12
provides these as P4 instances they're
00:07:14
essentially high performance machines
00:07:16
built for AI workloads and there is
00:07:18
three key things that make AI
00:07:20
infrastructure unique the first one is
00:07:23
computing power which we just mentioned
00:07:25
you need specialized GPU machines to
00:07:27
handle AI task efficiently the second
00:07:29
second one is network speed since AI
00:07:31
moves huge amounts of data around the
00:07:33
world you need extremely fast networks
00:07:35
to keep everything running smoothly and
00:07:37
finally storage AI works with massive
00:07:40
data sets we're talking about terabytes
00:07:42
or even petabytes this data needs to be
00:07:45
stored in data lakes and properly
00:07:47
prepared before AI can even use it now
00:07:49
as an AI Cloud engineer you'll design
00:07:51
these systems manage a GPU
00:07:52
infrastructure create secure data
00:07:54
pipelines whilst keeping costs under
00:07:56
control now that you understand the
00:07:58
foundation let's get to building your
00:08:00
first real AI project so for phase two
00:08:02
we're going to start a simple small
00:08:04
project that shows you how to piece
00:08:06
everything together terraform AWS and AI
00:08:09
now you might be wondering why we're
00:08:11
using terraform instead of clicking
00:08:12
around in the AWS console now as an
00:08:14
engineer you will never be clicking
00:08:16
around the console and writing
00:08:18
infrastructure because in the real world
00:08:19
you need to be able to deploy and update
00:08:21
your infrastructure reliably anyone that
00:08:23
tells you otherwise has no idea what
00:08:25
they're doing especially the ones on
00:08:26
YouTube Imagine manually setting up all
00:08:28
your AI systems by clicking buttons on
00:08:30
the AWS console 3 months later when you
00:08:32
need to recreate that same setup you
00:08:34
won't be able to recreate it as there's
00:08:36
no record of what you've done so please
00:08:38
please please do not spend any time
00:08:41
making infrastructure for a real project
00:08:43
using the AWS console I know there's
00:08:45
loads of people on YouTube showing you
00:08:47
the tutorials of how to build an ec2 and
00:08:50
you know what it's fine to get familiar
00:08:52
with the console but in the real world
00:08:54
when you're working on projects you will
00:08:56
not be clicking around and building
00:08:58
infrastructure with with the console
00:09:00
you'll be writing it as code
00:09:02
specifically with infrastructures code
00:09:04
so please please please don't watch any
00:09:07
videos that show you how to use the ads
00:09:09
console to build a cloud project it's
00:09:11
fine initially so you get familiarity
00:09:14
with that service but if you want to
00:09:16
build a real project for a portfolio you
00:09:18
have to use IAC so for our AI Cloud
00:09:21
project we'll spin up Cloud
00:09:22
infrastructure piece by piece using
00:09:24
terraform firstly we'll create VPC with
00:09:26
subnets and an internet gateway to have
00:09:28
access to the internet then we'll create
00:09:30
an S3 bucket for storage and set up your
00:09:32
IM roll with just enough permissions to
00:09:34
make things work with Amazon bedrock and
00:09:37
S3 Amazon Bedrock is aws's AI service
00:09:40
now as everything is defined in code you
00:09:42
can Version Control it and recreate it
00:09:44
exactly the same way every single time
00:09:47
now for this project we'll work with
00:09:48
some Simple Text data maybe a blog post
00:09:51
or user feedback logs you'll store all
00:09:53
of this data in S3 and use Amazon
00:09:55
Bedrock to process it this data will be
00:09:57
a knowledge base Bedrock will give you
00:09:59
access to powerful pre-trained AI models
00:10:01
through simple API calls think of it as
00:10:03
sending a request and then getting AI
00:10:05
response back no need to build an AI
00:10:07
from scratch we use Bedrock to summarize
00:10:09
the text or do some analysis now to
00:10:11
bring it all together you need to write
00:10:13
a python script that handles this
00:10:14
workflow uploading the text to S3 making
00:10:17
API calls to bedrock for processing and
00:10:19
then saving the results back to S3 and
00:10:21
because this is a real world setup will
00:10:24
Implement proper security controls
00:10:26
encrypting the s3e bucket and using the
00:10:28
least privileged am policy icies for all
00:10:30
of our Bedrock calls now when you finish
00:10:32
this project you have built your first
00:10:34
AI powered pipeline in the cloud a
00:10:36
production ready environment that
00:10:37
connects infrastructure storage AI
00:10:40
models and also your code and by the way
00:10:42
if you want to start building projects
00:10:43
today then I do want to tell you that
00:10:45
I'm running a special cohort right now
00:10:47
and we're taking in new students for my
00:10:49
cloud engineer Academy if you're ready
00:10:51
to take action just like Jay Martinez
00:10:52
who got laid off as a banker and got
00:10:54
Cloud hired in just a few months or Mac
00:10:57
who after joining my Academy landed a
00:10:59
systems engineer role at AWS then click
00:11:02
the link in the description and book a
00:11:03
call with my team but just know that
00:11:05
this isn't for everyone because it's not
00:11:07
possible for us to bring on you know 100
00:11:09
people at once because the spots are
00:11:11
limited and also the demand is really
00:11:13
high right now and cloud is booming so
00:11:15
we want to make sure we can actually
00:11:17
help the people that we bring on so if
00:11:19
you're interested go book a call and see
00:11:21
if you qualify for our special AI Cloud
00:11:23
cohort now before we dive into building
00:11:25
Advanced AI Cloud projects you need to
00:11:27
understand how modern AI systems are
00:11:29
structured so there are fre connected
00:11:31
layers that all work together to create
00:11:33
an AI application that people actually
00:11:35
use like chat GPT and understanding each
00:11:37
layer is important because they impact
00:11:39
one another if one layer isn't working
00:11:41
properly the whole system can break down
00:11:43
so number one we have the infrastructure
00:11:45
layer this is the foundation that
00:11:47
everything runs on the specialized GPU
00:11:49
servers high-speed networks and storage
00:11:51
systems that we discussed earlier
00:11:53
without solid infrastructure none of
00:11:55
these AI capabilities above it can
00:11:57
function properly number two the model
00:11:58
layer this is where all the AI
00:12:00
intelligence lifts the llms the large
00:12:02
language models or even custom models
00:12:04
built for specific business needs models
00:12:06
need mlop systems to automatically
00:12:08
maintain update them as new data comes
00:12:11
in they also use rag retrieval augmented
00:12:13
generation to access company data making
00:12:16
their responses more accurate and
00:12:18
relevant number three the application
00:12:20
layer this is what most people actually
00:12:22
interact with like when you use chat gbt
00:12:24
scrolling Netflix and you see their TV
00:12:26
show recommendations now while as a user
00:12:29
you never see the infrastructure or the
00:12:30
models underneath these applications can
00:12:33
only be as good as the layers supporting
00:12:36
them and these layers are all dependent
00:12:37
on each other for performance so if the
00:12:39
infrastructure is slow the applications
00:12:41
will be slow if the models are
00:12:42
maintained correctly the applications
00:12:44
won't give you a good response that's
00:12:46
why we need to understand and build each
00:12:48
layer carefully now you've already set
00:12:50
up a basic AI pipeline at the beginning
00:12:52
of phase one connecting infrastructure
00:12:54
storage and AI models it's time to take
00:12:56
things further but automating the entire
00:12:58
process ESS and this is where ml Ops
00:13:01
machine learning operations comes in
00:13:03
earlier we manually ran the workflow
00:13:05
uploading data to S3 calling AI models
00:13:08
through Amazon bedrock and saving the
00:13:10
results but imagine your data is
00:13:13
constantly updating like new customer
00:13:15
feedback or daily sales logs manually
00:13:17
handling this every time would be a
00:13:19
nightmare especially if the AI model
00:13:21
needs retraining as patterns in the data
00:13:24
starts to evolve with machine learning
00:13:25
operations or mlops we can automate the
00:13:28
entire life cycle so the new data
00:13:30
arrives in S3 the pipeline will
00:13:32
automatically kick things off then the
00:13:34
data gets prepared the pipeline cleans
00:13:36
transforms or formats the data as needed
00:13:39
then the model is updated if necessary
00:13:41
the pipeline will retrain the model
00:13:43
using the new data then the updated
00:13:45
version is tested and automatically
00:13:48
rolled out this means you don't have to
00:13:49
manually manage updates it just works
00:13:52
adapting to changes in real time now for
00:13:54
this project we'll build on what you
00:13:55
created in Phase 2 now you already have
00:13:57
the VPC the IM rolls and S3 storage
00:14:00
we'll expand on this by integrating a
00:14:02
sage maker pipeline to handle the
00:14:04
machine learning workflows and instead
00:14:05
of only calling the pre-train models
00:14:07
through bedrock you'll now trade and
00:14:09
deploy your own simple AI model the
00:14:11
pipeline will automatically trigger when
00:14:13
new data is uploaded to S3 retraining
00:14:15
and redeploying the model without any
00:14:17
manual input with your infrastructure in
00:14:19
place and an automated mlops pipeline
00:14:21
running we're now going to be making the
00:14:23
AI smarter by connecting it to business
00:14:25
data the next stage is all about rag
00:14:27
retrieval augmented generation we built
00:14:29
a pipeline using Amazon Bedrock to
00:14:31
process text Data with pre-trained
00:14:33
models and these models were powerful
00:14:35
but they didn't have any knowledge of
00:14:37
your business so you already have S3 set
00:14:39
up as a storage system but now you use
00:14:41
it to store your company's internal
00:14:43
documents policies and manuals customer
00:14:45
logs or any technical specification also
00:14:48
the pipeline that we made earlier in
00:14:49
phase two you know the one that process
00:14:51
text Data with Amazon Bedrock here you
00:14:54
enhance it by adding a vector database
00:14:56
using AWS open search to make the
00:14:57
documents searchable by AI a vector
00:15:00
database just means that your AI can now
00:15:02
find and retrieve specific information
00:15:04
from your data before using a model like
00:15:06
Claude to craft answers and here's how
00:15:08
we're going to build it in three simple
00:15:09
steps firstly you want to start by
00:15:11
taking your documents and turning them
00:15:13
into embeddings by the way you can just
00:15:15
generate documents in chat gbt for this
00:15:16
part then you want to load these
00:15:18
embeddings into open search index where
00:15:20
they'll be organized for efficient
00:15:21
retrieval then you want to connect open
00:15:23
search to your existing Bedrock based
00:15:25
pipeline when a query comes in the
00:15:27
system will search your data for
00:15:29
Relevant documents pass the results to a
00:15:30
language model to generate a accurate
00:15:32
context aware response now so far your
00:15:35
AI has been great at answering questions
00:15:37
and handling tasks but what if it could
00:15:39
work independently planning making
00:15:41
decisions and executing complex
00:15:43
workflows without constant oversight
00:15:45
this is where AI Agents come into play
00:15:47
AI agents combine the intelligence of
00:15:49
large language models with the ability
00:15:51
to interact with tools access apis and
00:15:54
take meaningful actions all
00:15:56
automatically they go beyond just chat
00:15:58
Bots and systems that can actually get
00:16:00
things done in this phase you'll build
00:16:02
an AI agent that automates the endtoend
00:16:04
business workflow for example you could
00:16:06
create an agent that pulls Market data
00:16:08
analyzes it for anomalies and sends a
00:16:10
slack alert with its findings building
00:16:12
on your existing Pipeline with this
00:16:14
agent will call apis to gather the
00:16:16
information it needs using python and a
00:16:18
connected llm it will process the data
00:16:20
to identify patterns anomalies and or
00:16:23
trends like finding underperforming
00:16:25
products or seasonal spikes it will then
00:16:27
generate detailed reports upload them to
00:16:29
S3 and then distribute them directly to
00:16:31
the team members or it can notify them
00:16:33
via email or slack to orchestrate this
00:16:36
you can use Amazon Bedrock agents this
00:16:38
enables your agent to work seamlessly
00:16:41
across different tasks that said while
00:16:43
these systems are incredibly powerful
00:16:45
they must also be secure and reliable so
00:16:47
you'll set up access controls and we use
00:16:49
IM rolls to control exactly what each
00:16:51
part of the system can and cannot do
00:16:53
setting specific permissions for who can
00:16:55
access data or run certain processes
00:16:57
everything your AI does gets tracked in
00:16:59
cloudwatch this means if something goes
00:17:01
wrong you can see exactly what happened
00:17:03
and fix it quickly and the reason why
00:17:05
this matters is because AI agents are
00:17:07
the future of every industry and this
00:17:09
project prepares you to work at the
00:17:11
intersection of cloud engineering Ai and
00:17:13
security and by combining AI
00:17:15
intelligence workflow Automation and
00:17:17
strong security practices you are
00:17:18
creating a production ready solution
00:17:21
that aligns with real business needs and
00:17:23
drives real business value now phase
00:17:25
three is taking your new skills to the
00:17:27
real world so here is is where I see the
00:17:29
biggest opportunities and most
00:17:31
importantly how to actually capture them
00:17:33
every business is drowning in data but
00:17:35
very few know how to use it and this is
00:17:38
your leverage Point while other
00:17:40
Engineers are focused on building
00:17:41
complex AI systems the real value is in
00:17:44
something simpler but more impactful
00:17:46
helping businesses understand and use
00:17:49
their data to create personalized
00:17:51
experiences for their customers take an
00:17:53
e-commerce website most sites still show
00:17:55
the same products for everyone imagine
00:17:57
Building A system that learns from each
00:17:59
customer's behavior and interactions and
00:18:01
creates a completely personalized
00:18:03
shopping experience or look at customer
00:18:06
service small businesses are struggling
00:18:08
with basic support tickets building an
00:18:10
AI system that can handle common
00:18:12
questions and root complex issues to the
00:18:14
right people that's immediately saving
00:18:16
the money and improving the customer
00:18:18
satisfaction and the key is that you
00:18:20
don't pitch the technology you pitch the
00:18:23
business impact and the business value
00:18:25
nobody cares about your rack systems
00:18:27
your Cloud infrastructure or your
00:18:28
machine learning pipelines instead tell
00:18:30
them that you can help them understand
00:18:32
why their customers are leaving before
00:18:35
let's say September and help them
00:18:37
predict it for next year or predict
00:18:39
which products will sell out for the
00:18:40
next season or which products will be
00:18:42
super popular for Black Friday maybe you
00:18:44
can ultimate all their support to handle
00:18:46
80% of the tickets automatically now
00:18:48
while everyone else is using generic AI
00:18:50
models you can build specific systems
00:18:52
tailored to business data and their
00:18:54
needs and that's where I see the biggest
00:18:56
opportunity right now is in customizing
00:18:58
ation of businesses and their services
00:19:01
and this is where the real value is in
00:19:03
2024 this is where the biggest paychecks
00:19:05
will come from and if you want to learn
00:19:07
AWS then click this video right here to
00:19:09
watch my brand new course that I've
00:19:10
released which takes you from complete
00:19:12
beginner to expert a toz and everything
00:19:15
that you need to know
