Forem: Mai Nishitani

FinOps with AI using Snowflake across multi-clouds

Mai Nishitani — Tue, 07 Oct 2025 09:42:57 +0000

I’ve noticed a trend, and apparently AWS has too.

Multi-cloud. There, I said it. The M word was strictly forbidden while I was at AWS, now there’s documentation around it.

Like it or not, the world is moving to multi-cloud and lots of change.

Change meaning a lot of uncertainty in terms of economy and the tech landscape with the rapid advancement of AI.

As organisations move towards cutting costs, while still rapidly experimenting with AI, how would you as a builder show your value to your manager and their bosses?

The answer is easy.
Make money appear in front of them.

But, I don’t work in Sales or Marketing you might say.

You don’t have to.

The answer lies in cost optimisation across your cloud workloads to "make money appear from no-where".

At AWS I helped customers find where some of the levers were to help save costs e.g. right-sizing, using Savings Plans and Reserved Instances etc.

There are multiple dashboards that provide customers visibility into what their spend is on AWS.

Great! But many customers were stuck with what they needed to actually action with this data.

With most Enterprise customers being multi-cloud, do they need to do the same thing across multiple clouds?

Yes, and that meant more time spent on analysing individual dashboards and compiling a list of actions for each Cloud Service Provider.

After achieving the Linux Foundation’s FinOps Certified Practitioner and FOCUS Analyst certifications I had a lightbulb moment.

What if we can convert all the CUR (Cost and Usage Report) data from each Cloud Provider, convert it to a FOCUS format data and analyse it?

As it happened, I was curious around playing around with Snowflake and jumped on a “Zero to Snowflake in 90 mins” hands-on labs.

Well done on a really interactive session Snowflake! I've delivered and participated in many hands-on lab sessions in my time both virtual and in-person, but that was one of the most engaging sessions I've attended.

I looked around for FOCUS sample datasets that contain multiple Cloud Service Providers which I found on GitHub here.

Then I was looking around for any Snowflake tutorials that I can run cost optimisation queries for, but I couldn’t find any.

It re-ignited being able to build something quickly and intuitively on Snowflake using AWS behind the scenes on a free Snowflake account.

So initially I was having a look at what’s possible with the visualisations via SQL statements.

I was happy with it initially but then I realised, Snowflake dashboards are quite simple and are not for full fledged visualisation like Tableau or even Amazon QuickSight.

So I went one step further by creating a Streamlit app via the in-built Snowflake feature and had a FinOps chat interface.

This makes it more intuitive as you can select quick actions, or ask it specific questions using the same data in the Data Warehouse using Cortex.

Here are some of the highlights of the Streamlit app.

One of the first questions is around modernisation to help with cost optimisation. A lot of organisations that I worked with still had a lot of tech debt, with monolithic apps running on Amazon EC2s. Modernising them into microservices gives customers the benefit of optimising costs as you move to containers and serverless services.

Some people spend the weekends going outside, but since it was rainy here I figured I’d spend my time building something useful.

It’s all about showcasing your value to your organisation by thinking big and inventing and simplifying, based on customer needs.

Whether that customer is an internal customer or external, it doesn't matter.

You should be continuously improving by keeping up to date with your technical skills, just like AI is by adding data to the vector database in a RAG model as your "context".

This blog is about FinOps with AI tooling that helps with visibility.

If you want to learn more about FinOps for AI workloads across multiple environments, have a look at this awesome overview https://www.finops.org/topic/finops-for-ai/

Using drones for FinOps with AWS Agentic AI - Part 2

Mai Nishitani — Sat, 27 Sep 2025 13:35:52 +0000

Here’s a common scenario, you have an application running which has some sort of AI features.

This could be a custom application that you or the team built or even a third party application or a combination of these.
Does this mean that your work is now complete?

Not quite… There is more to this.

Of-course as part of operational excellence or day 2 activities, you need to look at maintaining your AI workload.

But what exactly constitutes an AI workload especially as there are different ways to consume it than just your standard compute and storage and pay as you go pricing of cloud-managed services that we’re used to?

And why is the finance team suddenly yelling at me for overstepping my allocated budget?

As an Engineer, does it even matter? Isn't it enough if you just build some cool stuff?

I used to think exactly like that, as the cost stuff was the "boring part". Every time cost or governance was mentioned, I tuned out, because it wasn't like you got to talk about the new shiny tool that I could use.

Since going over to a more senior role, I realised that there is more to just cool technical tools.

Especially with the influx of new AI tooling after a good 2 years since the whole generative AI craze, I could now see many teams within organisations struggling with the spike in Proof of Concept costs that didn't get productionised. Where in fact, was the Return On Investment?

And where would you typically start in this case?

Start small and define your AI workload for your use case and make this visible.

At least that’s what the FinOps organisation says about AI workloads.

For a comprehensive approach of how to start making your costs visible across your AI workloads, this article from the official FinOps org has some practical tips.

You may be wondering if you've read this far, I get the FinOps part, but what's that got to do with drones?

Now you may have noticed that this is part 2 of the blog and part 1 is available here which talks more about where drones fit in.

As a refresher, the app that was built then was around using images collected by drones for predictive maintenance with the help of multi-modal LLMs orchestrated by AI agents on AWS.

Let’s say you need to now add some FinOps features for this workload, albeit in this case across AWS.

In reality, it may be a little more complex, as you may have systems outside of 1 cloud hyperscaler with multiple SaaS and other systems on other hyperscalers or even on-premises.

I added the FinOps functionality, using the FOCUS v1.2 standard.

Why is it important that I used the FOCUS standard? It's because AI workloads can be spread across multi-cloud and multi-environment (SaaS, Data centers) and is unlike what we're used to before in your typical workload around applications which tend to include, compute, storage and other pay-as-you-go type services.

I’ve asked Kiro which is AWS’s spec-driven code generation tool to help me add the FinOps functionality, within the Brisbane Bridge Dashboard front end page.

The initial design was this logical architecture. In order to get real-time AWS Cost and Usage Report (CUR) data, there was a lot of moving parts that was inefficient.

Kiro did try to get real CUR data from my AWS account but failed despite a number of steers.

While pondering about the next step to resolve this issue, I remembered how you can integrate any MCP servers with Kiro.

I configured the official AWS Cost Explorer MCP server with Kiro to take out the guess work.

Super useful to have the following tools which align to the first step in FinOps to gain visibility of your AI workloads via tags.

Let's take a look at how this panned out when I asked Amazon's Kiro to help me add this functionality on the dashboard.

You can see in the dashboard it gives you an overview of what the overall spend is, and what the Return On Investment is and the cost savings are based on factoring the AI workload in this case all on AWS which includes the following for this specific app: (YMMV - Your Mileage May Vary)

1. Generative AI/LLMs/Agents - Amazon Bedrock including Bedrock Agents - Claude and Nova models Tokens in and out, other pay per use costs
2. Storage associated with this workload - Amazon S3
3. Compute associated with this workload - AWS Lambda functions
4. Databases associated with this workload - AWS DynamoDB
5. Other AWS-managed services costs - Amazon API Gateway, Amazon CloudFront
6. Other Networking costs - data transfer costs over 100GB

In a real-life scenario, you would also need to factor in any SaaS usage and licensing costs e.g. other LLMs from other hyperscalers, HuggingFace, Snowflake or Databricks for your data or other AI/ML, observability, security and governance related systems related to this workload too.

From an orchestration standpoint, do you use Apache Airflow, Kubeflow, other CI/CD costs.

Maybe additional compute might be used like some containers with Amazon EKS or ECS. There might also be some Spot capacity used for batch processing.

What about if you incorporate some additional AI services like Amazon Textract, Amazon Comprehend etc? That's another part that needs to be considered.

Not to mention if using RAG or CAG, your vector database costs including Amazon Bedrock Knowledgebases or Amazon OpenSearch for the Vector Database component in RAG.

For CAG, you might be looking at Amazon ElastiCache or Amazon API Gateway.

How would you calculate Fine-tuning costs?

What if you replicate data across multiple AWS regions starting with Amazon S3? What if your AI workload is so critical that it's active-active?

How about if there's spikes in usage with anomalies where the token counts go up and the context size increases unexpectedly? Are you prepared for the cost blowout?

If you think that's too much for one cloud environment, we also can't forget that many have AI workloads spanning multiple-clouds and on-premises as well.

Some may even have most of their core data on-premises and may want to run their inference close to their data. Now do we need to consider egress costs from the hyperscalers out?

This may sound very overwhelming and the first step is to start with the use case and agree on which components should be included for a small use case.

This includes all of these personas according to the FinOps article.

Getting alignment with the above teams is the most difficult part of the process, solving the technical part of the challenge is easy.

Keen to hear how you’ve solved how to get alignment on what an “AI workload” means for your use case and organisation?

Using drones for good with AWS Agentic AI - Part 1

Mai Nishitani — Sun, 21 Sep 2025 00:19:03 +0000

Drones are in the news lately - most notably Cardi B with breaking the Guiness World Records with the most number of albums delivered via drones in an hour.

Drones can be used for even better things like helping the community from disasters.

Speaking of community, I had the pleasure of presenting at AWS Community Day and AWS User Group in Brisbane of recent weeks.

In the sessions, I talked about how you can use drones for a good purpose, before, during and after a disaster.

At NTT, we provide drones to customers via our e-Drone Technology team which deal with 2 types of drones.

One being the aerial photography Anafi drones and inspection Skydio drones.

The drones can be programmed with autonomous flight and 360 degree object avoidance.

It can also do thermal and LiDAR imaging and collect comprehensive video and still imagery.

We also build our own drones for agricultural usage, such as inspection of crops for effective use of land.

Before disasters

As for helping before disasters, we have the predictive maintenance angle.

At NTT, we help customers do predictive maintenance across their hydraulic power plants.

Who uses solar panels for their home or business?
Although not specific to a disaster and more around a predictive maintenance angle, this use case where drones can detect faulty panels would be useful for Australian households and businesses.

During a disaster

For help during a disaster, we’re able to assist firefighters to see where the fire has spread to and help locating people that might be at risk of danger and drones can extinguish the fire.

After a disaster

As for post-disaster events, drones are able to help with navigating through difficult terrain where it’s unsafe for people and vehicles to get to.

Even restore internet connectivity by flying over landslides by re-laying fibre optic cables.

You can see our drones in action here on our YouTube channel.

NTT e-Drone Technology - YouTube

NTTグループのドローン専業会社です。NTT e-Drone Technology is an NTT Group company.

youtube.com

In my session, I talked about how we can use AWS services to help scale our existing ML image analysis model and enhance it with Amazon Bedrock. The idea is to create a dashboard application on React hosted on AWS Amplify from the front-end user perspective that shows the range of Brisbane-based bridges that are in scope for the analysis.

Each of the bridges are scored, low, medium or high risk according to a combination of human in the loop (bridge inspectors) plus Amazon Bedrock and Anthropic Claude models.

I gave Claude the prompt “You are a bridge inspector. Assign a bridge risk status based on the data we have about the bridge”.

The data in this case is produced by the proprietary ML model NTT eDrones runs to produce the images with cracks and rust super imposed over the top plus the CSV and CAD data.

The key is that we have thousands of hours of high quality drone imagery that we’ve captured over the past around infrastructure such as bridges and piers.

I also used some help with Kiro to come up with baseline user stories for Bridge Inspectors and Engineers, Maintenance Co-ordinators and IT Administrators and generate code on an existing git repository.

Kiro is Amazon's spec-driven code generation tool which gathers requirements on your behalf.

I managed to create a Kiro coaster with punch needling technique. The eyes are a bit wonky, I'm a punch needling noob - this is an MVP.

I also happened to be a participant at AWS Community Day New Zealand (Aoteroa) in Wellington for an hour or so in time to catch the last keynote by AWS Developer Advocate, Donnie Prakoso.

The future is bright with the AWS New Zealand region just opening - I saw a lot of engaged Developers and Technologists in the Community Day session.
Thanks for the AWSome mouse pad - great idea.

Here's my actual session from a different AWS Community Day in Brisbane, Australia.

If you want to know what went on behind the scenes at AWS Community Day, take a look at my previous blog.

After the session, I spoke to an Amazonian who mentioned that I should look into incorporating Amazon Nova models and agents into the mix as an idea.

So I did, now with this high-level design.

What does the multi-modal agent AI capability look like in this demo solution?

AI components used

After that discussion I thought about why we might need to use AI agents for this solution.
Does it actually make sense to even use agents in this case?

We could potentially use it for multi-modal analysis.

There’s text analysis and the human-in-the-loop approach and also for comparing LLMs.

Amazon Bedrock + Claude 3: What I started off with

Think of Claude 3 as your trusted Bridge Engineer who never gets tired.

We feed it inspection data and get back proper engineering reports, based on the detailed prompt around being in the shoes of a Bridge Engineer.
Even the best AI needs a sanity check from real engineers, so we will always need input from Engineers in the field manually inspecting bridges.

We made sure there was a traffic light system of risk (High, Medium, Low) classifications with confidence scores for each bridge.

It also gives you approximate figures for repairs, which is useful for budget planning in the short, medium and long term.

Amazon Nova + AgentCore

I’ve used Amazon Nova model with AgentCore for multi-modal analysis.
In addition, as a way to see whether the Nova models stack up to Claude models.

By incorporating agents into the data analysis (image and text) it does so more efficiently than the original solution calling Claude via Amazon Bedrock.

In terms of image processing, Nova Vision models can spot cracks in bridge photos better than most untrained humans.

Nova Text models can analyse CSV files of measurements of these defects and find patterns.

Nova Reasoning models combine everything into actionable recommendations based on the data collected.

AgentCore Orchestration

There are multiple agents working together to orchestrate the different LLMs.
It also manages the workflow for complex analysis processes.
Built-in fallbacks when AgentCore doesn’t work, it gracefully falls back to direct Amazon Bedrock Agent access.

Multi-Modal Analysis

Image Processing (Nova Vision)

It does everything that the Claude models do on Bedrock as per below.

Automatically analysing any structural damage from the image, classifying the risk level, tracking how damage develops over time and location mapping.

Data Analytics (Nova Text)

It uses CSV files as input for finding trends in measurement data that humans would miss.
It also tracks how things change between inspection rounds for each bridge, important for predictive maintenance.
Data validation is part of the analytics as it catches measurement errors before they cause problems.

Synthesis Engine (Nova Reasoning)

This is where everything comes together based on the data outputted from the text and multi-modal models.

Combining visual and measurement data for comprehensive analysis.
Gives you short, medium and long-term recommendations.
In terms of costs, it helps prioritise repairs based on risk vs. cost.

Event-driven architecture approach

This is where using AWS Lambda helped with triggering events when the file lands in the Amazon S3 bucket.

This is so that it can trigger bridge analysis events with real-time status updates.

From a risk perspective, it was important to alert any changes to the status for co-ordinating safety responses.

Real-Time Update System

From a customer perspective you can leverage some of the 5G, LTE network connectivity that our drones have.

At the Brisbane AWS User Group session, I talked about how Amazon's Project Kuiper could help with Satellite connectivity across Australia's rural areas when it's launched in 2026. It may even help with network coverage for drones working in rural areas.

We still need to consider the data transfer costs of real-time vs batch image processing. If we need to say setup a private LTE network, there will be costs associated with setting up the networking hardware required for this.

From a front-end integration perspective, I used WebSocket integration for live updates of analysis results and alerting for multiple users and profiles.

Intelligent Caching System

In terms of optimising performance, caching helps improve response times and reduces API costs with minimising repeated calls to Amazon Bedrock.

Fallback Mechanisms

Graceful degradation is part of the solution with structured fallback from AgentCore to direct Nova model access.

Monitoring and Observability

Using Amazon CloudWatch for real-time metric tracking for overall service health, for time, costs and error rates.

Edge Processing capability

This solution can handle on-device AI analysis if running ML on the edge.

In addition, the solution can continue its autonomous operations with offline capabilities, with default batch processing if required.

Security and Compliance

Security is job zero for us, and that's what I used to say back at AWS. This solution is no exception with secure processing with encryption and access controls.

I used Amazon CloudWatch and CloudTrail for comprehensive logging for regulatory compliance and quality assurance.

Also least-privileges using IAM role-based permissions with appropriate data access levels.

Let me walk through the front-end and the explanation around the ability to switch from Claude to Nova models.

What would the impact look like for the solution in the future?

We can look to integrate Real-time IoT sensor data integration for continuous monitoring capabilities.

In addition, we can look at GIS Integration with Spatial analysis and geographic information.

We can look at make the job easier for Bridge Inspectors out in the field with real-time guidance and support.

Also, from a regulatory compliance perspective, we can automate generation of summaries for compliance reporting.

How drones can help for good

By combining drones with AI analysis, we can provide communities globally a more efficient disaster planning and response, before, during and post the event.

The integration of Amazon Bedrock's Claude models with Amazon Nova's multi-modal agents creates an advanced level of intelligence in analysing infrastructure such as bridges and piers.

This solution provides actionable insights, cost estimates, and prioritised response plans that enable communities to recover faster and more effectively.

This solution demonstrates how emerging technologies can be used for social good, providing critical capabilities when communities need them most.

Stay tuned for part 2 of this blog where I dive into the FinOps and advanced cost analysis dashboard components.

If you've been working in this space, I'm sure you can imagine your Chief Finance Officer saying “tell me what the Return on Investment is with this AI workload”?

Read more about it here on part 2 - FinOps for AI workloads + drones.

AWS Community Day Australia - Behind the scenes as a speaker and participant

Mai Nishitani — Sat, 16 Aug 2025 11:32:48 +0000

2018 - The first ever AWS Community Day in Sydney, Australia

I was presenting about AWS IoT and Amazon Sumerian during my AWS customer days. I was so nervous, and did pretty much the full rehearsal at the stage with all my memorised lines.
I remember the "community spirit" and how everyone including volunteers, AWS partners and AWS staff worked as one to get this event off the ground. It changed the trajectory of my career, as it was the catalyst to me joining AWS to give back to the community and mentioned this in my interview at AWS. I met many AWS Solutions Architects (now ex-AWS) and now AWS Heroes and Community Builders there.

2019 - AWS Community Day in Melbourne

This event was bigger than the 1st one and exactly 4 weeks before I joined AWS. I remember awkwardly telling my boss at the time who also traveled to attend this event, that I've decided to join AWS as an Associate Solutions Architect. What was memorable from that event was that Kris was talking at this event (unfortunately I can't recall what topic it was), and there was a panel session with diverse women speaking about their experience in tech which was relatable.

Fast forward to March-April 2025

It all started with an announcement from Alan that this year’s community day after a hiatus of many years (2177 days!), is going to be in Brisbane.

I got super excited and sent through quite a few suggestions for the speaker expression of interest back in March/April and was lucky to be selected.

May-August 2025

My talk for revolutionising drones with Amazon Bedrock got selected!

Speaker prep (insert Rocky theme here)

After that, it was the usual presentation prep cycle that I did back at AWS.
Come up with a rough idea for my slides (really basic dot points to begin with).
Start creating proper visuals for my slides.
Think about where and what demo to put in and build that demo.
Record that demo, re-record it, edit to a reasonable length.
Start writing dot points around the talk track.
Actually practice out aloud, scrap the talk track, start again.
Keep practicing eventually without any speaker notes with just the visuals - rinse and repeat.

2 weeks out, Stephen sent me a Slack message to ask if I wanted to be on the Community Day panel as well.
Of-course, my answer was yes!

I set on the way to Brisbane and because I somehow have nightmares of being late to meetings and conferences, I stayed overnight in Brisbane so I can be there at the Brisbane Convention Center nice and early.

I met up with the AWS Community Day crew including volunteers and speakers for the pre-party. It was great to see some old and new friends in the community. I left early so I can go and practice some more.

T-0 - AWS Community Day

Woke up at 5:30AM, got ready, practiced my talk again and the panel responses. Headed on my way across the bridge to get a Nodo Donut for breakfast.

Turned up around 7:30AM at the Brisbane Exhibition Centre.
There was a flurry of activity but patiently waiting for testing my laptop on the stage and get logistics around microphones etc.

The day started off with a blast with the keynote with Alan Blockley and his ode to Brisbane and the rest of Australia with what makes a community special.

What I got was that it’s not just about knowledge and skills on AWS but the spirit of helping others so that this event is successful.
I can hear the passion in his voice around bringing this community day to Brisbane and all the effort that the team of volunteers and sponsors put in.

And what a day! Big thanks to Alan Blockley, Michael Kingsley, Arjen, Lucy, Jade, Stephen, Mark P, countless AWS partners and sponsors for all the blood sweat and tears! I can imagine that there must have been so much planning that went on after-hours to get this off the ground.

Jason’s session talking about the Ibis aka Bin Chicken being the mascot of the Brisbane User Group got me in stitches!

Despite being in Brisbane myself, I forgot about the iconic Bin Chicken = AWS User Group.

Also the BBS from back in the day, brought back some memories from way back with racking up 100s of dollars on my parent’s phone bills on a dial-up modem.

Then it was the Community Day panel session with Dmytro, Jason and I with Stephen as the moderator.

Stephen, who is also an AWS Hero did a great job of moderating our panel session with the questions to the right person at the right time and did it with a natural flow. I commend him on continuing to give back to the community by giving guidance and opportunities for speakers across diverse backgrounds within the AWS Community.

I was really interested in hearing from Dmytro about his journey from Ukraine to Australia to leading the AWS Gold Coast User Group as an introvert.
The introvert part, I can really relate to as well and I’m sure others will too.

I’ve been to the GC User Group myself, and I highly recommend to drop by even if you’re visiting on holidays.

How Jason was able to build out the Community from AWS Builders and how they can be awarded the golden jacket through official means.

Soon after, it was my turn on the stage.
I was glad that I was able to articulate my main points around how drones can revolutionise disaster recovery in the real sense using AWS.
My emphasis was around “drones for good” for the community as sometimes it can literally be weaponised.
Looks like Kiro was the hot topic for the day with other speakers. I made a note to myself that I’ll change up the content a bit for any future iterations of the talk.

Nevertheless, I was relieved that I finished my panel and session, so I was much more relaxed after that.

In between the sessions, I was busy catching up with current and former AWS folks, and talked about the good old days.
It brought back a lot of nostalgia around when we all rallied around and built something for customers, purely because we care.
We still have that startup mentality inside and that special AWS LP of “Think Big” that excites us even for those have since left AWS.

Speaking of which, it was Kris’s talk about “Chat GPT did not write this talk”. It was a very thought-provoking session, that made me re-think the use of gen AI in certain settings. How can we use AI for good while trying to be authentic?

What makes humans be our messy, imperfect selves?
Somehow John Legend’s song comes into my mind “with your perfect imperfections…”.

The last session was with Derek and Mike. It started off with an intro video about the 3 little bears which was super creative.
Then there were Minecraft references. I didn’t expect snippets of news articles in between with different opinions for and against and in-between the gen AI bandwagon.

I won’t give too many spoilers, but you’re going to have to check out the YouTube video when it comes out.
I said to both Derek and Mike, you guys need to do more sessions together because I loved it!

The closing keynote by Alan left us all warm and fuzzy thanking all of us, including the volunteers, partners and speakers and the community for making this day the best day ever - after 18 months of planning.

I’m keen to help for the next community day as a volunteer next time.
Let me know, and I’ll travel to help this AWSome crew.

We always say find your tribe… Looks like I found mine.
You might find yours if you go to your local AWS Community Day or User Group.

Scaling and troubleshooting Amazon EKS just got easier with MCP on Anthropic Claude

Mai Nishitani — Mon, 05 May 2025 03:12:59 +0000

I remember back in my days of being a Systems Administrator, I had lots of fun building a monitoring server from scratch with Nagios. Anyone old enough to remember it? Looks like Nagios is still alive and kicking these days too.

What that involved every time there was a cryptic Linux related error, I had to do the “man” command or —help to start off with. Then running verbose mode to try to debug without much success. Then searching the internet for any clues, usually on Stack overflow at that time and it was either “Duplicate question, closing” or “I solved the problem!” Without any details on how they solved it. It was frustrating and I remember either trying to do a temp workaround of some sort or give up configuring that particular feature.

More than 15 years later - things have gotten much much easier in some ways but also complex.

The move from VMs to containers and going from tightly coupled monoliths to micro services.
Was micro services meant to make your life as a Platform/DevOps Engineer simple?

Maybe - but there are still complexities on how to continue being operationally excellent and keeping up with the dreaded Kubernetes version upgrades.
This works if you’ve got a handful of non-critical applications running. What about in an enterprise scenario where you do have critical workloads running and more than a few of them?

I’ve lived through troubleshooting Kubernetes with a customer - rather, trying to upgrade their Kubernetes version and making it work with their applications on non-prod. We were frantically working to fix this as a new Kubernetes version was going to be force rolled out in their prod environment for security reasons, and the customer was so far behind in versions.

Basically it was a team of people trying all the different kubectl commands under the sun to try and troubleshoot what’s broken - without prior knowledge of the workload. Somehow, we were able to avoid the imminent disaster in time, but it wasn’t ideal.

After that incident, I was continually on the search around how do we make Kubernetes operations simple for customers.

This coincided with the launch of Amazon EKS Auto Mode.
What is EKS Auto Mode? I would say a better, improved version of AWS Fargate for EKS.

On Fargate for EKS, there were quite a few limitations, including the ability to not be able to run daemonsets and other networking and storage considerations.

EKS Auto Mode brings best practices of Amazon EKS into one place without most of these limitations.
For example, EKS Auto Mode keeps your Kubernetes cluster, nodes and other components up to date with the latest patches up to a 21 day window.

This still means that your organisation needs to make sure that your applications running on EKS need to be tested and ready for the next Kubernetes version.

I started playing around with Auto Mode per the AWS documentation and then a thought came to my mind.
 What if you could run connect an MCP server to Amazon EKS Auto Mode?

MCP is a protocol that helps standardise communication with LLMs.
According to Anthropic, it’s like a USB C port to connect data sources and tools.
Basically, the aim is to reduce operational overhead of development.

Why would anyone want to do that with Amazon EKS?

Well wouldn’t that help with troubleshooting endlessly with kubectl until the break of dawn? (I shouldn’t have given up on my rap career)

Then I stumbled into this GitHub repo. Shout out to Alexi-led for an awesome project.

So I decided to give it a try. It wasn’t difficult to get started at all.

I was impressed by how it could automatically run the kubectl commands via the Claude Desktop UI.
No more frantically looking up the kubectl commands and switches on the CNCF webpages.

Wait… does that mean CNCF certifications like CKA (Certified Kubernetes Administrator) and CKAD(Certified Kubernetes Application Developer) is going to be obsolete?

I wished at that time we had something like an MCP server that could interrogate your EKS workloads and ask it questions in natural language. It sure would have made the troubleshooting much easier!

So here’s a video of me asking questions about my current Amazon EKS setup.

It dives into great detail around the current configuration, including the number of pods, deployments and services etc.
I was then curious around whether it was running Auto-Mode or EC2s behind the scenes. I think in this case, it was undecided as it said it was Auto-Mode and some type of managed services but it had instances starting with i- . The correct answer is yes this is Auto-Mode, as Karpenter has been installed out of the box as with other automations.

I then asked it to scale the amount of nodes to test out Karpenter. It did this seamlessly and I was taken by surprise. I know Karpenter is meant to scale automatically without you having to trigger or input the scaling, I just wanted to see if Claude knew what it needed to do. And of-course testing the scaling back down gracefully was satisfying as well.

So imagine being able to ask Claude to troubleshoot especially if you’ve exhausted all options and it’s 2AM in the morning and your brain isn’t working.

Wouldn’t it ease some stress off your day to day life as an Engineer?

How to build a voice 2 voice Severance bot with Amazon Nova Sonic

Mai Nishitani — Thu, 24 Apr 2025 07:40:22 +0000

Hands-up who’s a fan of Severance?

I am, and I’ll try to write this without spoilers.

The show reminds me of corporate life and how you can have spurts of fun in between.
It’s a mix of comedy and mystery - but not as cringe as “The Office” it’s subtle but there are funny parts between all the serious parts.

So what’s this got to do with a voice 2 voice bot?

I wanted to quickly run an experiment for fun, a Proof of Concept to see how quickly you can get started with using the new Amazon Bedrock FM, Amazon Nova Sonic.

What’s unique about Amazon Nova Sonic? It takes your voice as Base64 and uses it to output voice.

Think of it as a game changer for contact center and chat bot applications as well as generating your personalised assistant.

Why is Amazon Nova Sonic so exciting?

For those that have been building on AWS for a long time, in order to build any interactive voice bot, you might have used services like Amazon Lex to build out chatbot responses. I remember at least back in the day, you had to predict how the conversation might go with “intents” and “slots”.

Then you had to use Amazon Polly which is a text-to-voice service to convert the resulting text to voice. Think of Amazon Polly like the voice that powers Amazon Echo devices (Alexa! Play a funny joke).

For memories sake, here’s my experiment that I did a few years back with Amazon Sumerian which was deprecated but awesome service which helped you create Avatars and interactive bots.

The intent of this video was to help mentor the new interns at AWS.

So I started looking at other blogs, notably this one here.
I started off trying to build the front-end with a bit of genAI-ception with Amazon Q Developer CLI.
Amazon Q Developer CLI helps with not only code-generation but actually deploying the AWS services required to build the backend and frontend.

But it all got a bit complex as I wanted the backend code to be Python.

At the moment, the Nova SDK is experimental with Python - there wasn’t much data that Q CLI could use as a baseline despite referring it to the GitHub page.

So the quickest way to getting this up and running was running the Amazon Nova Sonic workshop that AWSome Amazonians have written.

It involved a bit of troubleshooting in terms of setting the Python version to version 3.12 and running the requirements.txt and making sure that the Python server.py was running first before running npm start.

In addition, you need to configure the Amazon Bedrock Knowledge Base and generate a KB ID. The Knowledge Base uses Amazon OpenSearch in the background as a vector database that will help provide specific internal knowledge that the LLM can use.

In this case, I used Amazon Nova Lite for retrieving the content from the Knowledge Base.

I originally wanted the Nova Sonic to talk about my top 10 Karaoke songs, and although it worked while testing this on Amazon Bedrock Knowledge Bases, it didn’t seem to like that question and referred to general karaoke top 10 songs not on the KB when I asked Matthew who is the voice of Nova Sonic.

Instead I ingested the Severence Wiki page about Lumon Industries.
Notice that when asking, it talked about Lumen Industries and lighting - it could be my pronunciation, but I wished there was a feature to block referring to outside knowledge and only the KB. I may need to look into that with KB settings or Bedrock Guardrails.

Here’s the link to the Amazon Nova Sonic Voice 2 Voice workshop.

There is a GitHub repo and the readme that I referred to when self-running the workshop.

Going back to Severance, I can imagine all of the characters being peculiar, working at a tech company like AWS.

Mark is definitely an Account Manager, he’s reasonable and operates great with customers.

Helly is definitely in a C-level role, maybe a CIO - she has a presence that fills the room.

Irv is the Solutions Architect - he’s great at analysing things and articulating it in front of people.

Dylan is the Developer - of-course he’s bothered by Mr Milchick breaking his flow.

And what about Mr Milchick?
Probably a Solutions Architect Manager who likes to let loose every now and then.

Who’s your favorite Severance character and what would you build with Amazon Nova Sonic?

Put it in the comments below!

The Kubernetes platform engineering north star - How to convince your boss that an internal platform accelerates business

Mai Nishitani — Thu, 24 Apr 2025 03:14:59 +0000

So you want to convince your boss that having a platform engineering strategy for your IDP or Internal Developer Portal is important.
First of all, what is an IDP?

An Internal Developer Platform (IDP) is a centralised, self-service system within an organisation that gives software developers everything they need to build and deploy applications without dealing with the underlying infrastructure complexity.
Think of it like a company-specific virtual vending machine (service catalogue) that provides standardised templates, workflows, and resources so developers can focus on writing code. This is so they don’t have to mess around with configuring servers, setting up databases, or managing CI/CD pipelines.

IDPs typically include monitoring and observability, APIs, and automation tools that handle repetitive technical work. Bad news for rogue developers who run shadow IT, but an IDP can enforce company security standards.

Conversely, one of the many benefits of the IDP is that it helps with reducing the cognitive load for developers.
But where do you start with building one or if you already have an IDP, whether it’s built on best practices?

We have so much information these days, and you might not have time to distill all the information to make sense of it. Not to mention, there’s already a few issues piling up that need your immediate attention.

You and your team might be dreaming about that single platform that rules them all, where all your developers will willingly use your platform to deploy all the infrastructure they need.
While your platform operates seamlessly, without you being woken up at odd hours in the morning.

In reality, it’s not that easy - Kubernetes can be complicated if you let it. There are so many ways to get your platform wrong, especially if you’re not keeping up with the version and dependency upgrades which inevitably comes around every few months.
If you don’t have standardisation around how you’re going to operate this IDP, then people in your team wouldn’t know how to fix it when it’s broken.

I’ve seen many customers struggling with getting started with 1. Convincing their boss around the benefits of building an IDP and 2. Continue to advocate for the platform to gain use. Unfortunately, if you build it, they (developers) won’t come running to it. Not to mention 3. How do you ensure you have operational excellence built in so you don’t suffer from dreading Kubernetes upgrades and suffer catastrophic application failures because of it.

Haofei Feng and I co-wrote this document around how to start thinking strategically on platform engineering and improving the maturity around your Kubernetes model. Although we wrote this with Amazon EKS in mind, the same principles can be applied to any Kubernetes platform you’re using.

I’ve been channeling Gregor Hohpe and his platform strategy book throughout this process. I highly recommend having a read of his books!

Why Platform Engineering Matters

Why should your organisation care about platform engineering in the first place?
The short answer: Developer productivity and organisational agility.
The longer answer is that a well-designed internal platform lets your developers focus on writing code instead of wrestling with infrastructure.
It provides consistent, self-service access to the cloud-native capabilities they need, enhances developer flow, and ensures security and compliance by default.
Yes you need to think about security and compliance in your organisation, no matter how boring this might sound.
While you’re at it, it doesn’t hurt to become friends with the security team. They’ll give you some great pointers to look out for if your idea needs approval internally.

The Platform Engineering Vision

A successful platform engineering vision has several key components according to the CNCF Platforms White Paper:

Platform as a product: It should be designed with your end-users in mind, likely developers focusing on building common use cases across products.
Focus on user experience: It should meet users where they are, offering multiple interfaces (GUIs, APIs, CLIs).
Documentation and onboarding: We’re all guilty for this, you gotta have documentation. Make sure you not only write how to use it but keep your docs updated frequently. Ever encountered a time where someone in your team took some time off and couldn’t fix an issue because it wasn’t documented anywhere? Additionally, it’s important you do some internal training and mentoring to help onboard the users that are going to use your platform. This also acts as a bit of an internal marketing exercise too.
Self-service capabilities: Users should be able to request and receive capabilities via self-service. What does that mean? Ultimately you want developers not having to mess around with infrastructure to make things work. Wouldn’t it be great if they can get a single page web application vended to them like a virtual vending machine? Imagine not having to deal with troubleshooting permission errors to make something work.
Reduced cognitive load: Hide complexity and implementation details from users. It’s what Gordon Ramsey used to say, Keep It Stupidly Simple (KISS). Would you want to deal with a heavily complicated platform that has all the features and patterns that you can imagine from the start? Or start simple with solid functionality and solicit new features and feedback from your users?
Optional and composable: Teams should be able to use only the parts they need and not be forced to use all the features. The more de-coupled your services are, the less blast radius. Let’s move away from the tightly coupled, monolithic approach.
Secure by default: Compliance by design and validation based on standards. Don’t have a standard? Maybe start building one. Here’s a good baseline to refer to.

The north star architecture

What does a target state architecture that incorporates modern cloud-native practices look like?

Infrastructure as Code: Terraform and Crossplane for declarative infrastructure
GitOps-Driven Automation: Argo CD and Argo Workflows for deployment and orchestration
Developer Self-Service: Backstage as a developer portal to centralise workflows
Scalable & Flexible Architecture: Kubernetes as the core platform with various optimisation strategies
Platform Governance: Integrated with IAM, security policies, and compliance controls

Source diagram: https://github.com/cnoe-io/reference-implementation-aws

Using the CNCF Platform Engineering Maturity Model

Great! You have all of the above. Can we officially call our IDP mature enough?
You can assess your platform using a four level maturity model based on the CNCF Platform Engineering Maturity framework:

Level 1 - Provisional: Ad-hoc, reactive responses to team needs
Level 2 - Operational: Dedicated platform teams, standard interfaces, centralised tracking
Level 3 - Scalable: Self-service solutions, centrally orchestrated capabilities, standard processes
Level 4 - Optimising: Intrinsic pull from users, integrated services, quantitative and qualitative measurement The above model evaluates maturity across five areas:
Investment: How staff and funds are allocated
Adoption: How users discover and use platform capabilities
Interfaces: How users interact with the platform
Operations: How platform capabilities are planned and maintained
Measurement: How feedback is gathered and incorporated

You can read more around the maturity model here.

Path to platform maturity

What do you need to do if your platform is not there yet?
There’s a three-phase transformation approach:

Phase 1: Solidify the Foundation

Establish security controls and boundaries
Define operating model and RACI
Create IaC blueprints and DevOps pipelines
Document common consumption patterns
Develop comprehensive documentation

Phase 2: Scale and Enhance

Implement advanced self-service features
Develop an internal developer portal
Improve governance with preventive and detective controls
Collect and analyse user feedback
Provide advanced training for platform consumers

Phase 3: Optimise and Innovate

Foster experimentation culture
Hire and develop engineering talent
Form cross-functional teams for inner-sourcing models
Leverage emerging technologies like generative AI
Build robust feedback loops

You can see what this is all about in this YouTube video here about Platform Engineering with Amazon EKS.

How to talk to your boss about this

If you're trying to convince leadership to invest in improving your platform or building one from scratch, here are some talking points:

Focus on business outcomes: When I was an Infrastructure Engineer (what you’ll call a Platform/DevOps engineer these days) I struggled to tie the “business outcomes” to the tech that I was talking about. Think beyond the scope of what you do day to day. What would your team, your department and your entire organisation benefit from long-term with this platform? You can mention that this platform can help accelerate developer productivity, improve operational efficiency, enhance organisational agility, and be able to scale and scale back down when not in use, reducing costs. Here’s some more data points that you can reference in your pitch.
Show the adoption pattern: Initially, platform adoption is slow and there might be resistance within the internal teams. You need to show the developer teams the value that the platform provides by internal presentations. What if you’re afraid of running presentations or even speaking up in front of your colleagues? That used to be me, 10 years ago. I used to get really nervous and my heart would race every time I spoke within a large group of people. You know what helped? Toastmasters. There’s one near your neighbourhood or even at your workplace if you’re lucky. If not, there’s online toastmasters which you can join in. Start there, and work your way on speaking more in front of people. Don’t worry, you got this! Once you’re able to do some internal promotion of your platform, your users will be intrigued. Once they start using it, then it starts to provide value by creating a steeper adoption curve.
Highlight the cost of not investing: What we call the “opportunity cost”. Without a platform strategy, teams will duplicate efforts, security will be inconsistent, and developer productivity will suffer. Who’s seen different CI/CD tooling used by one team of developers vs another? The old tech adage of “if it ain’t broke don’t touch it” will lead to more technical debt.
Present a phased approach: You don't need to do everything at once. Otherwise you’ll fall into the trap of analysis paralysis. Start with having a solid baseline, then scale, then optimise.
Use the maturity model as a benchmark: Continuously check where your organisation currently stands and show the practical steps needed to advance to the next level. If you can gamify going to the next level and make it fun, the better for platform adoption!

Here’s a great way to capture your boss’s attention around Platform Engineering. Thanks to Frank Fan who is a Senior Specialist Container Solutions Architect for co-presenting and diving deep into this concept at AWS ANZ Summit.

Time to get building

The journey to platform excellence is a marathon, not a sprint. Yes we know, you want to get something up and running and want to impress your boss with results.
With a clear vision, a maturity model, and a practical roadmap you can refer to, you can build a platform that your developers rave about.
Not only that, your security teams will be impressed (wait, that would never happen. Security teams always seem to be grumpy), and your boss will be glad that you can help them deliver business value faster.

Remember, the goal isn't to reach the highest maturity level – it's about finding the right balance for your organisation's needs.
What's your current platform maturity level?
What challenges are you facing in your platform engineering journey?
I'd love to hear your thoughts and experiences in the comments below!

Author bio:

Haofei Feng (Dev.to user: fldhmily63319)
Senior Cloud Architect - Professional Services - AWS

Haofei is an innovative strategic technology leader with 19 years’ experience delivering impactful consulting, architecture, and managed services across the full project lifecycle—from strategic advisory and presales influence to hands-on execution and operational excellence. Renowned for aligning business objectives with advanced cloud, data, and AI solutions across regulated and complex industries. Proven ability to communicate complex technical concepts to executive stakeholders, build consensus, shape organizational vision and deliver transformation programs that drive measurable business value at scale. Recognized for authoring and presenting strategic whitepapers guiding operating model and maturity frameworks to enable sustainable business transformation.

Mai Nishitani
Director of Enterprise Architecture - NTT Data

Mai is a former-AWS Solutions Architect who is passionate about experimenting with new things.

How you can achieve all AWS certifications and get the golden jacket - Part 2 - How to study

Mai Nishitani — Sun, 13 Apr 2025 07:32:45 +0000

Now you’re motivated to actually work towards getting the golden jacket after reading Part 1 of this blog.

This blog will help you start preparing for the 12x AWS certifications (as of now).

How do I know what study method works? Well I’ve tried and tested this method myself as with many people that I worked with at AWS.

Working at AWS it was another level of busy day to day - so I had to maximise my study time without feeling too overwhelmed with too much information.

I know there are plenty of other AWS certification study material out there, but these are the courses that worked for me and my former colleagues at AWS that passed the exams.

First point of call - Official AWS study resources

I used to use A Cloud Guru (now Pluralsight) back in the day and some of the free practice exam pdfs that I could get from the AWS official certification page. Things have changed since then, and I find the AWS official training more than enough to learn and pass my exams.

Check out the latest list of AWS certifications by clicking on each badge to get to the latest and official AWS study resources from there.

It should take you to recommended exam courses on AWS Skill Builder which is the official AWS online training portal.

You might be thinking, I’ve already been studying for [insert AWS exam here] for [insert weeks/months] here. How do I know if I’m ready?

Also, you don’t want to spend more money on study resources because you already used a few dollars on it.

Here’s some great news for you. No matter what AWS exam you’re studying for, you can revise using the AWS Training Live on Twitch shows to see if you can understand what the hosts are talking about.

Best of all, the best things in life are free [insert Janet Jackson and Luther Vandross track] thanks to the hard work that the AWS team led by AWS Principal Developer Advocate, Aaron Hunter.

As a former AWS Training Live host while at AWS - we’re dedicated to interacting live with our online audience running through some exclusive questions all while having fun.

Here's a screenshot from the session around AWS Power Hour - AWS DevOps Professional and fellow hosts, May Kyaw and Harrison Sherwin.

Level 1 - Starting on AWS? Do AWS Cloud Practitioner

If you're just starting on AWS, it might be less of a heavy-lift to do the AWS Cloud Practitioner course and build your confidence.

I’ve heard some say that you can start on AWS Solutions Architect Associate if you're already technical.

What can sometimes happen is that due to the SA Associate exam covering a wide range of topics, people get busy at work and start to lose focus with studying and keep postponing the exam.

So I say start small with Cloud Practitioner and celebrate the small wins!

For a fun way to learn, you can also use a gamified version of it on AWS Cloud Quest.

I would highly recommend doing this, as you'll have fun using the lab (which is currently free as far as I know) without having to rack up costs in your own AWS account.

If you're already heavily into the AI/ML space, should you do the AI Practitioner exam? Of-course!

I found that a large chunk of the exam questions seems to have general AI/ML knowledge in there.

On the flip side, if you’re not in that space yet, I’d start off with AWS Cloud Practitioner.

Level 2 - The AWS Associate certifications

Now that you’ve got the Cloud Practitioner exam under your belt, how would you study for AWS Solutions Architect Associate and SysOps Associate and ML Engineer Associate?

Same pattern of studying as Cloud Practitioner but more wider and in-depth.

It's all about building that muscle of pattern recognition.
Some of the questions are quite long-winded, and in some cases only a few words different.

The key is to read the question and see what is it asking for. Usually the last sentence will be a clue to what you’re looking for in the answers. What’s the most cost-efficient, performant solution? etc.

I scan the answers to see what's different between the responses. There should be at least 1 that you can identify that's definitely wrong. You can cross that one out in your mind.

Weirdly enough, out of the AWS Associate exams, I found the Data Engineer Associate exam quite difficult.
It was even more difficult than the Data Analytics Specialty exam that since got deprecated.

Again, use Skill Builder on AWS, I highly recommend the Exam preparation courses which the basic version is free of charge (as of now).

Level 3 - What about the Specialty exams?

As the name implies the Specialty exam dives deep into the topic of choice.
This is where you need to dive deep into the topic and again pick up more detailed patterns that you’ve built up using your knowledge in the Associate certifications.

The hardest AWS exam in my opinion is the AWS Network Specialty exam. For this one, you'll need to study more than the usual method that I have articulated above.

For the Network Specialty exam, getting to know the hybrid pattern (connecting on-premise to AWS) is super important and take a look at the recommended whitepapers from the AWS certification page.

I was under the impression that my past networking certification (Cisco Certified Network Associate) would help me, but I was wrong.
This exam is focused on AWS-specific networking patterns and only some generic networking concepts.

Speaking of which, there were some IP subnetting type questions (but not as much as I imagined) so I suggest understanding those fundamentals.

If you don’t want to lose motivation, I would leave the Advanced Networking until the very last.
Even after doing all your Specialty and the 2x Professional certifications.
Why? Every person that I’ve spoken to, even seasoned AWS staff agree that this one is the most difficult exam out there.

Level 4 - The professional certifications

The Solutions Architect Professional and the DevOps Professional certifications... are they difficult?
Yes in terms of the number and depth of questions plus the lack of time during the exam.

Time management is key here for these exams. The questions I found are the longest in any of the exams. This made me read the questions a few times in order to fully understand what it was asking for.

What makes the professional certifications easier would be to read through the Well Architected Framework whitepaper many times. Highlight some of your learnings. When it comes to the exam, you'll be able to build on what you’ve learned so far in the foundational, associate and even specialty exams.

Gotta catch ‘em all - Sequencing of study - YMMV!

This is what worked for me is the following sequence.

You might be asking, wait how come there’s 15 of them?
As you can see, there’s been quite a few certifications that have been deprecated since.

I also took into mind what works best for me when re-certifying. Your Mileage May Vary as you would have different areas that you’re strong in.

i.e. Get the foundational and associate exams done first so you can auto-renew when getting the professional certifications.

There are some that are not in that order, only because these exams came out at a later stage after I got the golden jacket.

AWS Certified Solutions Architect Associate (at that time there was no Cloud Practitioner)
AWS Certified SysOps Administrator - Associate
AWS Certified Cloud Practitioner
AWS Certified Developer - Associate
AWS Certified Solutions Architect - Professional
AWS Certified DevOps Engineer - Professional
AWS Certified Data Analytics Specialty (old exam - deprecated)
AWS Certified Database - Specialty (old exam - deprecated)
AWS Certified Security - Specialty
AWS Certified Machine Learning - Specialty
AWS Certified Advanced Networking - Specialty
AWS SAP on AWS - Specialty (old exam - deprecated) >> Received golden jacket here as I finished all the Generally Available AWS certifications at that time
AWS Certified Data Engineer - Associate (new exam - beta at the time)
AWS Certified AI Practitioner - (new exam - beta at the time)
AWS Certified Machine Learning Engineer - Associate (new exam - beta at the time)

Other resources

What if you want more material to study with?

I sometimes supplemented my studying with the following resources so I can be fully prepared with more patterns that I can have up my sleeve.

There are plenty of other resources for self-study like Jon Bonso's Tutorial Dojo and Stephan Maarek's courses.
If you or your organisation has an account on Udemy, you can checkout courses from both Jon, Stephan and others from there too.

Also shout-out to AWS Community Builder Lucy Wang with her Tech with Lucy YouTube series who has a bunch of videos around certification.

Speaking of AWS Community Builders…. Part 3 of this blog will showcase the Golden Jacket hall of famers who happen to be AWS Community Builders or Heroes.

Let us know in the comments how you’re traveling on the journey to your golden jacket!

How you can achieve all AWS certifications and get the golden jacket - Part 1

Mai Nishitani — Sat, 05 Apr 2025 06:00:04 +0000

So, you're thinking about getting all AWS certified. It might be a long road ahead to get the golden jacket, but now is the best time as you only have to get 12 AWS certifications (as at 5 April 2025).

What is the golden jacket? It's awarded to anyone that passes all the AWS certifications that are generally available at the time. It doesn't include any certifications in beta.

You also not only need to pass them, but need to have it active and not expired. This does get tricky as you'll need to re-certify in time if your certification is up for renewal.

Why get AWS certified?

I always get asked the question, why did you decide to get all 14 AWS certifications? What motivated me?

The answer is...
the same reason people decide to run a marathon.
If you've never run long distances before, the first 5k are going to be painful.
You're going to be out of breath, you might even get a stitch on the side while running and your legs are going to hurt.
You might even think about quitting before you finish running because it's too hard.

I also consider myself a life-long learner and on the quest for knowledge.

Certifications are like running a marathon

After conquering the first 5k, it's going to get easier, bit by bit.
Then you might start entering your first official 5k race, you get through that race, maybe even get a personal best time. You feel that accomplishment and you're super keen to run the next longer race. Maybe a 10k or even a half marathon?

You can apply the same concept for starting out with AWS Cloud Practitioner. Especially if you're just starting out in cloud or even tech, the concepts might be difficult to grasp at first. But once you pass your first exam, you'll be hooked on the adrenaline rush just like running.

Now that you've finished the first 5k, what about the next 10k? That's your AWS Solutions Architect certification.

My first AWS certification

I started out with the AWS Solutions Architect certification back 10+ years ago, where AWS Cloud Practitioner certification didn't even exist. As a Systems Administrator(these days you would call the role a DevOps/Platform/Cloud Engineer), cloud was new and we were very much tied to on-premises workloads. Even the CIO and CEO was scared of the cloud as it was talked about as something that they should be afraid of, due to potential security impacts to their current environment.

I wish that I could travel back in time and tell the CIO and CEO what I know now. The collection of best practices, The Well Architected Framework goes through "security in the cloud" vs "security of the cloud". As customers we still need to be responsible for what we run on the Cloud. Ultimately, we still need to have an internal security policy that we base our security configurations in the cloud. Most times, organisations that I've spoken to over the years don't have one or if they do, the cloud team are not aware of it. But I digress...

As a Systems Administrator back then, I was happy running Virtual Machines on VMware in our own private cloud. I hadn't heard of what cloud was until that fateful day...

An application vendor came in and he was flustered, he told me that our CIO was worried about using an application running in the Cloud due to potential security risks. So he was asked to re-deploy this app in our private cloud on VMware instead. I can tell he wasn't happy and I was tasked to create him a LAMP stack ASAP.
The app was deployed successfully and was up and running in no time.

I did get curious around cloud at this point though. What is it about cloud that people were so fearful of? There was only one way to find out... getting certified!

Changing roles from Systems Administrator to Solutions Architect

Back then, I used A Cloud Guru (now Pluralsight) and free AWS training content to start studying. Getting hands-on definitely helped, especially getting to know the foundations of compute and storage that I was familiar with. Amazon EC2, Amazon S3, AWS IAM at the time was an easy concept that I could grasp based on experience. Perhaps it helped that there weren't 200+ services back then.

Once I passed my exam, I started to think whether there were opportunities elsewhere. Internally, the organisation was nowhere close to using cloud at that time. So I took a bet and started a contracting job for another organisation that required cloud skills.

During my role there as a Solutions Architect, I was able to run demos and deliver results quickly using my newly acquired skills.
Now I could match my newly found skills with my day to day role.

There were times that were challenging, people would ask why should you use cloud vs on-premises to run applications.
I still feel that's a common theme across any organisation even now, where people are afraid of change.

As a recent example, the use of generative AI across organisations. Is your company embracing it investing in generative AI projects with guardrails, or steering clear of any mention of generative AI?

Ask yourself, are you happy with your current organisation and the technical direction? If you feel that you're a pioneer, but cannot use your skills in your current organisation, you need to re-think about what's best for your growth.

So what's the best way to pass all the AWS certifications?

As Andy Jassy would say:

There's no compression algorithim for experience

You can learn quickly if you start building on AWS at work.

But what if you're not experienced or not using AWS in your day to day role? Are you doomed? Not quite...

With or without experience on AWS, you will need to study and get hands on. It's the best way to learn and everyone has to start somewhere.

Stay tuned for part 2 of this blog where I go through the study resources that I used to get AWS certified...