CGreciano's Blog

Loved the Music of The Lion King, but not it's AI-generated scenes

2026-04-16T00:00:00Z

This past weekend I attended a concert titled “The Music of The Lion King” in Helsinki. I had been looking forward to the event for weeks, since The Lion King holds a very special place in my heart. Ever since my childhood, I have loved every bit of the movie, and it remains to this day one of my favorite movies of all time. A big part of that is its majestic music, with artists such as Elton John and Hans Zimmer. You might then understand why I was so excited to come to this concert. It’s not The Lion King Musical, but I was still hyped.

The concert delivered when it came to the performers and the music. The singers were amazing, their voice really powerful. The musicians and orchestra brought tears to my eyes on several occasions. The familiar music touched my heart. The concert also included a homage to Africa and some interesting facts surrounding The Lion King’s impact worldwide. Heck, we even got Shakira’s “Waka Waka” shoehorned in, and we were invited to join and enjoy the show. 😀 I was delighted to get a picture with the performers after the concert.

But there was a stain. During the performance, we had several screens showing images and animations related to the songs and The Lion King. And let me emphasize the “related” word. They were not the original scenes from the movie. They probably didn’t have the rights from Disney to show those in the event. Instead, we got clearly AI-generated videos of live-action animals resembling those of The Lion King. Sometimes they were OK to watch, other times the color palette and the soulless AI-slop were extremely distracting. During the interpretation of the Stampede theme, the wildebeests appeared to be carelessly treading instead of running for their lives. The hyenas seemed to have an ocean inside their mouths as they drooled excessively. And the animal bodies often had disproportionate limbs and features. The coloring in the scenes was also overbloated.

All of it naturally led to the key question: “would it have been better to NOT have any content at all over having AI-generated content?”. It’s a tough question to answer, and it will of course vary from person to person. In my humble opinion, I would have preferred no visual nor animated content at all than this AI-generated slop. And I know that some people worked on putting that animation together, I know that I’m complaining about someone else’s work. But the fact is that sometimes I had to close my eyes to more fully enjoy the concert and not get distracted. And what bothers me is that this drop in quality is just going to permeate in every form of media moving forward. I don’t mind quality content made by AI. But I do mind AI-slop being portrayed as quality content.

Have you been in any events recently where AI really turned you off? What is the best way, do you think, of showing dissatisfaction towards this trend (assuming we don’t want to go to the extreme of avoiding quality events just because they throw AI into them)?

Run your Local AI in less than 15 minutes! (no tech skills required)

2026-03-30T00:00:00Z

It’s 2026 and hence you’re most likely already using AI tools such as ChatGPT, Claude, or Gemini. You’re probably also worried about the rising costs of using these AI tools and there’s good reason to be: vendors of these tools are rising prices and/or introducing ads to try to be profitable. If you are entrenched on AI workflows and becoming dependent on a certain vendor, what will you do if the vendor spikes their prices to something you can no longer afford?

The antidote for your worries is using local AI! By running a free and open-source AI model in your machine and hardware, you can say goodbye to monthly subscriptions and ads! You might be scared of trying to install local AI on your computer and think it’s a complicated setup, but I assure you, it is surprisingly simple to do so, as I will show you in this blog post. If you’re here just for the instructions, jump directly to the Step-by-step installation of your local AI section. Before that section I want to first explain the tradeoffs between using local AI and cloud AI and then briefly present the AI stack of tools we will use to deploy AI models locally.

Tradeoffs between local AI and cloud/SaaS AI

SaaS stands for “Software-as-a-Service”: many of the online apps you use, like Gmail, Netflix, Spotify… are examples of SaaS. ChatGPT, Claude, etc are also SaaS, AI SaaS to be specific. Let’s be very clear on the pros and cons of running your local AI versus using an AI SaaS, i.e. a vendor’s AI hosted in the cloud. If you run your local AI, you are hosting the AI model(s) in your own hardware, which means:

👍 No subscription fees and no ads. Compare using Netflix or having video files locally on your computer: if you have the video file of your favorite movie in your machine, you can reproduce it as many times as you want from now until forever, without having to consume ads or paying a subscription to Netflix.
👍 No internet dependency → low latency (super fast transmission), can work offline. The AI is in your hardware. Your requests to it and its responses to you travel almost instantly within your machine, whereas accessing an AI SaaS means traversing the internet. This advantage is really noticeable if your internet connection is unstable and/or slow.
- ‼ Note that this advantage does NOT guarantee that your local AI will always be faster than an AI SaaS! If your local AI needs to “warm up”, if it’s bottlenecked by resources in your machine, or if you’re using a slow reasoning model, that extra waiting time might be higher than waiting for responses and requests to traverse the internet.
👍 Much more data privacy. This is very important for many users. While you can disable cloud AI from using your data to train their models, your data is still traversing the internet and is still being stored in the vendor’s servers. What’s worse, cloud AI models can learn enough about you and store sensitive stuff about you in their long-term memory. What guarantee do you have that vendors won’t sell your data to third parties for targeted ads (or worse things)? Notice how social media knows you scarily well? By using a local AI, your data stays in your machine, and thus stays private.

At the same time:

👎 Your hardware puts limits on how powerful your local AI can be. Some AI models can be very demanding and resource intensive. Some require multiple GPUs if you want to self-host them, which are expensive (you, like most people, might only have your laptop and nothing else). Think how a cinema most likely has better sound, lighting, and atmosphere for watching movies than your local living room with your run-of-the-mill TV (but you gotta pay cinema tickets if you want to use it).
👎 Related to the above: AI SaaS are recommended if you use AI a lot in mobile devices. Smartphones and tablets generally have much less performant hardware than computers and laptops, so running local AIs on mobile devices is tricky (you usually have to settle for SLMs, i.e. Small Language Models, as opposed to LLMs or Large Language Models). You can of course set up a local server in your home with a self-hosted AI and have your mobile devices at home access that local server. But that gets into home-lab territory and I promised to keep things simple for non-techies in this blog post.
👎 Some AI vendors have excellent AI tools that go beyond simple chat interfaces. Stuff like NotebookLM, Claude Code, Claude Cowork… setting alternatives to these products locally is not trivial. If you use them a lot, it makes sense to pay for them, but if you mostly just chat with AI models, then you can most likely get away with local AI!

Local AI Stack we will install

We will only install a few programs to run AI locally. If you’re non-technical, you might not have heard of these yet, so let me briefly introduce you to them:

Ollama: this is a backend service that runs quietly in the background. With Ollama you can download open-source AI models easily, and once downloaded they are exposed and available for use.
Open WebUI: this is a free and open-source webapp that acts as a frontend to the AI models exposed via Ollama. It provides the chat interface that loads in your internet browser, which is really similar to the web pages of ChatGPT, Claude, etc.
Docker: many techies like me love Docker containers. Think of them like bounded and sealed boxes where you can isolate and automate certain programs and apps so that they don’t interfere with other programs and apps in your machine. It’s like putting ice cream into a small container so that eating it won’t get messy (as opposed to ice cream in cones, which CAN get messy!). We will install a Docker container that comes bundled and integrated with both OpenWebUI and Ollama, greatly simplifying the installation process.

Image source: Mega Cone Creamery

This proposed stack is really simple, but there are many ways and alternatives to run AI locally. If you’re interested, I will suggest some alternatives later in this blog post.

Step-by-step installation of your local AI

Download Docker Desktop (the version that works in your OS)
Open Docker Desktop, then open a Terminal within Docker Desktop (it should be on the bottom right corner of Docker Desktop, click on “Terminal”, and grant permissions to open one)

Open this link in your browser (it’s Open WebUI’s official GitHub, focused on the section where you can install Open WebUI bundled with Ollama via a single docker command).
Copy the command to run a Docker container with Open WebUI bundled with Ollama. You most likely should use “CPU only” (but feel free to experiment with “GPU support” if you know what you’re doing). If using “CPU only”, the command is: docker run -d -p 3000:8080 -v ollama:/root/.ollama -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:ollama
Paste that command in the Terminal of your Docker Desktop and hit enter. This will install Ollama and Open WebUI via Docker, which might take a while (several GBs of download).
Once it’s ready, you should see the open-webui container running in Docker Desktop (a green dot indicates it’s running). Notice that you can always stop, restart, or delete this container (which contains your local AI) via Docker Desktop (stop ⏹ and play ⏯ buttons to the right)

Open WebUI is ready to be loaded in your browser! Click on the “Port(s)” link in Docker Desktop for the container (should say 3000:8080), or navigate to http://localhost:3000
Create an admin login for Open WebUI (don’t worry, this login is needed only for local use, and it stays local).

Once created, login as an admin into your local Open WebUI, you will be greeted with a pop-up, which you can close. You will then see a chat interface similar to ChatGPT and other cloud AIs you might have used.
Before chatting with an AI model, we must first download it! And now comes the million-dollar question: which model should you download and use locally? Well, even though I will suggest a model in a moment, the annoying answer is “it depends”. This is like choosing which T-shirt to buy: often you need to go to the store and try on several different T-shirts before you find the one that fits you, feels comfortable, and you like how it looks. You can explore tons of different open-source models in Ollama’s website and see which one best fits within your machine.

Photo by Ashita Mata on Unsplash

All of that said, I suggest you start with Meta’s Llama 3.2. This is a smaller model that does a good enough job at the reasoning tasks that many people use ChatGPT for (summarizing text, drafting emails…), and is likely to run well in most modern laptops. If you have older or less powerful hardware, you might need to look for an even smaller model (you will know this if Llama 3.2 runs sluggishly or not at all). If Llama 3.2 runs super smoothly, you can explore more powerful models (Llama 3.2 is far from cutting-edge at this point, I just wanted to suggest a simple model to start off with)
To install Llama 3.2 in your computer, click on your Admin account name within Open WebUI (bottom-left corner in Open WebUI) > then go to “Admin Panel” > go to “Settings” > go to “Models” > click on “Manage” > enter llama3.2 in the field for pulling models from Ollama > click the “Pull Model” button > wait for the model to download (will take a few minutes)

Once the model is downloaded, you should see it under “Models” (you might not see it immediately, but you can wait, or refresh the page, or leave and come back to “Models”)

Now that our model is downloaded, we can chat with it! Go to “New Chat” on the top-left corner, and talk with your local AI!

Alternative stacks/tools for local AI

While the Ollama + Open WebUI stack we have used for Local AI is really straight forward to install and use, there are alternatives out there:

LM Studio: Best for those who want a visual “App Store” for local AI models. It’s a polished app that looks like a professional chat interface.
Jan: Open-source alternative to LM Studio which focuses on being a 100% offline personal assistant. If you’re a privacy-conscious user who wants a clean, simple UI, give it a try.

Wrapping up

I hope I convinced you to give local AI a try. Running local AI is much easier than what most people think it is, and it will definitely be worth it once cloud AI becomes too expensive and annoying. There is no need to vendor-lock yourself to AI tools!

I’d love to hear from you if my blog post inspired you to try out local AI and you liked it! And even if you disliked it, you can still reach out and tell me why! Good luck in your endeavors!

I passed AWS Machine Learning — Specialty (MLS-C01)

2026-03-17T00:00:00Z

As I explained in a previous blog post, when AWS announced their new Generative AI Developer Professional certification (AIP-C01), they also informed that they would be discontinuing the Machine Learning Specialty certification (MLS-C01). It’s going away soon (31 March 2026 is the last day one can take it), so I made an effort and decided to get it before it’s no longer possible. Last week I took the exam and passed! 🥳

Reasons to take this discontinued certification

Why get this certification if it’s going away? I had several reasons to do so:

The cert is still valid for 3 whole years. I won’t be able to renew it once it expires, but for the next 3 years the cert is valid and is a testament that I know about ML in AWS.
It was my first ever AWS Specialty certification exam and I was curious to probe its difficulty.
I had already passed the AWS Machine Learning Engineer Associate (MLA-C01) and the AWS Generative AI Developer Professional (AIP-C01) certs before. This ML Specialty certification had a massive overlap with the content from MLA-C01 (about ~85% of the content), so I barely had to learn new things to pass ML Specialty, just review a bunch of material.
ML Specialty asks you generic data science and ML questions that are vendor agnostic and not only AWS-related questions, so in a way I also got generic data and ML knowledge validated.

Impressions and Results

This ML Specialty exam felt slightly harder than MLA-C01, and much easier than AIP-C01 (to be fair, I did take AIP-C01 during its beta period, which makes it extra challenging). The results also seem to corroborate my feelings, I achieved quite a nice score while I barely passed AIP-C01 two months ago:

All in all I’m happy I got this cert, it was quite a low-hanging fruit for me since I already had done other AI certs for AWS. Unlike for the other AWS certs that I hold, I won’t be releasing study notes or flashcards for ML Specialty, since soon nobody will be able to take the exam anyway. But if you’re curious and want to tackle the challenge within the next two weeks, refer to my study notes and flashcards for MLA-C01, as I said 85% of the material is the same (just the focus of the exams is slightly different).

ML Specialty focuses exclusively on traditional ML and deep learning (DL) and does not validate your knowledge of modern Generative AI. I understand why it gets discontinued since GenAI is all the rage right now. Traditional ML and DL appear to be so much more niche these days in the industry. Still, I’m saddened by this trend. Generative AI got where it got thanks to traditional ML and DL. There are still plenty of problems in the industry that are much better suited for ML/DL than for GenAI, yet so many companies and projects are trying to shoehorn an LLM where it’s not optimal. It was also quite unique to have an AWS cert that asked you plenty of vendor agnostic questions, the new AI certs pretty much quiz you only on their AI products… Oh well. Hopefully we get to see this kind of cert in the future from AWS again!

A tribute to the GOAT of MTG Kai Budde

2026-03-07T00:00:00Z

I was honored to play against Kai Budde in June 2024

In June 2024, I participated in Magic: the Gathering’s (MTG) Pro Tour (PT) held in Amsterdam. While I had been playing MTG as a hobby for 6 years before then, it was still unbelievable to me that I had reached the top level of competition of this card game, yet there I was. For Round 2, I was paired with Kai Budde. Unless you’re into competitive MTG you probably don’t know the name. But this guy, nicknamed the German Juggernaut, holds tons of records in the MTG competitive scene, and is considered the Greatest of All Time (GOAT) by many. You can imagine my excitement and nervousness at the pairing.

MTG is a strategy game that has a heavy dose of variance, which means sometimes the less skilled player will beat the more skilled player. That’s exactly what happened that time: I got lucky and managed to eke out a 2-1 win vs Kai. The guy was super friendly and accepted to take a picture with me, allowing me to immortalize that unique highlight. If you’re into MTG and want to know more details of how that match went, they’re available in this video where I summarize my PT experience.

This was undoubtedly the highlight of the PT for me, but in a sad turn of events, Kai announced publicly the next day that he had what was likely terminal cancer. The MTG committee then announced that from now on, the prize for best MTG player of the year would be named the “Kai Budde Player of the Year”.

Kai passed away recently, and tribute videos and messages have flooded the MTG social scene. I have abandoned MTG since my participation in that PT, yet I felt compelled to write this blog post and dedicate some words to this MTG GOAT, who used competition to bring the best in him and others. His streak of winning several high-level tournaments in a row back in the early 2000s is legendary to this day. He will be missed in the MTG scene.

Thoughts on life and our given time

I am also reminded that human life is quite frail. We never know if we will live one more year, 10 more years, or 50. Hence, we should be mindful and careful in how we’re spending our time and life. Nothing wrong with enjoying life and indulging in pleasures, but are we foregoing opportunities that we will regret later? Are we living the life we want to live, or the life that society and others have imposed on us? What legacy do we want to leave behind when it’s our time to sleep and never wake up again? What and who do we value the most in our life, and are we spending the time and effort on those endeavors and relationships?

I am truly passionate about strategy games and technology. I find them fascinating. But I need to remind myself to not put them above my family relationships and my ethical values, for example. After finishing the MTG PT in June 2024, I came at a crossroad: get deeper into MTG, participate in more top-level tournaments, and potentially become a professional; or pursue other endeavors. The game was too addictive to me to be kept as a hobby. In the end, I decided to let go of it. After taking that decision, I started to use my free time to level up my skills in tech instead of playing MTG. I studied and passed AWS certifications, published and sold my study notes and flashcards, started my blog and a tech YouTube channel, built my personal website, and got up to date in the AI craze. Almost two years later, I’m enjoying a really engaging job position as AI engineer at Wave Company.

I might grind MTG again in the future. Maybe I don’t. No matter what I do, however, I will always remember that match I had the honor to play against the GOAT of MTG, as well as his passion for the game, for the thrill of competition, and for becoming the best he could be. Ruhe in Frieden, Kai!

Still trusting Notion with your data?

2026-01-31T00:00:00Z

Photo by Gwyn Hay on Unsplash (I added Notion logo and file icons)

A couple of days ago a Reddit user described a horror story where they lost tons of important data in Notion. Their Reddit post got a lot of traction and attention. It is just another reminder that we should care on where we store our important data, and who we entrust that data with. Cloud storage and cloud syncing are really convenient, but what if something goes wrong? Do you have a local backup of your most important data, files, and documents?

Disappointed with Notion

I came across Notion several years ago and I was stunned. A very easy UI to get started in, an opinionated formatting that simplifies workflow, and easy keyboard shortcuts for efficiency. Free to use, powerful, free and automatic sync between devices… Amazing product! And it must be doing something right, because its revenue continues to grow and grow with time.

However, for me Notion is a beautiful note-taking app with some useful database functionality, and THAT’S IT. I don’t believe it’s a good task manager nor a good productivity command center, despite what many Notion influencers might try to sell you. If it’s trying to compete with Microsoft 365 and Google Workspaces… it falls tremendously short at that. It has shoe-horned its AI to all its users, mostly slowing down loading times without bringing that much value.

Yes, loading times, a consequence of Notion’s biggest flaw in my opinion: your data is not local, it’s stored in their online servers. You can’t access Notion files in your device, and downloading your data in a different format is clunky: you can download your Notion pages as HTML, CSV, and Markdown for free (with limitations), but exporting your whole workspace as PDFs (the most convenient format for documents and printing) requires a Notion Business subscription. Notion introduced an offline mode a while ago, and it helps with loading times, but you still can’t access local files nor easily migrate to other note-taking apps.

Obsidian is local-first

The online nature of Notion made me look into Obsidian. Obsidian’s UI and out-of-the-box workflows are not as sexy as Notion’s, but it’s actually not that difficult to use (in essence, Obsidian is an enhanced text editor at its core). I can also see all my Obsidian data as Markdown files in my computer. I can easily migrate to another app if Obsidian stops being supported or something better comes along. I see my data. I control my data. I own my data. The convenience of cloud sync across devices is not free in Obsidian: you either pay a subscription to use their servers, or you set up your own synchronization across devices. It’s a trade-off I’m willing to accept.

I still use both Obsidian and Notion

Oftentimes Obsidian and Notion are pitched as competitors and for many use cases they are indeed competitors (“which is the best note-taking app?”). Myself, I use Obsidian much more than Notion these days: I use Obsidian for journaling, brainstorming, content creation, personal note taking… I still use Notion to create study notes that I want to share publicly though. Notion’s feature to create ad-hoc and free websites from any Notion page is super handy! But I fully understand that Notion could take my data hostage for whatever reason at some point. If that happens, however, I have PDF exports of my online Notion notes, so it wouldn’t be the end of the world. It is a bit of a pain having to pay the Notion Business subscription just to download PDF files of my study notes (and I also have to stitch the PDFs together into one larger PDF file), but I do get to offset that by selling my notes in PDF format (users who want to use my study notes for free can always study from the online notes in Notion).

Learn the data ownership lesson

If a product is wonderful, convenient, and free, investigate how they make their revenue. Understand the trade-offs involved compared to its competitors. Decide what you can and can’t live without. Be conscious about it, or you might live a nightmare scenario at some point. I myself can’t accept lots of important data not being available locally.

How I passed AWS AIP-C01 (Generative AI Developer Professional) in Beta

2026-01-19T00:00:00Z

2026 started in the best way possible for me: I took the Beta exam for the AWS Generative AI Developer — Professional certification (AWS AIP-C01) and I passed! 🥳 I’m one of the first 5000 people globally that passed this exam, so I also earned the Early Adopter badge. This is my first Professional-level certification with AWS, and it was no joke! Let’s take a look at what led me to take the exam, some details on the exam experience, and the study materials I used. As a bonus, I’m offering the flashcards I created during my study at the end of this blog post.

Background

I have previously made videos and blog posts about this certification, so one can trace how I decided to take the exam. When AWS announced this cert, I was honestly skeptical on whether I wanted to take it or not. Amazon has not been at the forefront of AI compared to its FAANG counterparts, and I had not yet attempted a Professional-level AWS cert. I also knew Beta exams are generally more difficult, since they’re longer and less refined (I did take the Foundational-level AIF-C01 while in Beta).

However, after giving the free sampler of 30 exam questions from Tutorials Dojo a try, and doing well at it, I realized I knew quite a lot of the material in scope already. I already held four AWS certifications: Cloud Practitioner (CLF-C02), AI Practitioner (AIF-C01), Solutions Architect Associate (SAA-C03), and Machine Learning Engineer Associate (MLA-C01). All four of them provide important foundational knowledge and I recommend taking them before attempting the GenAI Developer Professional cert.

Once the Beta released, I decided to attempt the exam in January, even if in Beta. I was motivated to take the exam early since I wanted to do it before changing jobs (I started at a new place mid-January), and also I wanted that shiny Early Adopter badge that AWS is only giving to the first 5000 people who pass the exam. I didn’t have a lot of time to prep during Christmas vacation, but I figured it could be just enough if I made an effort.

The exam experience

This was one of the toughest exam experiences in my life! It was expected, and it was indeed so. Almost 4 hours (with the 30 min ESL extension) for 85 questions, which meant 2-3 minutes per question. I ended the exam without a single minute to spare, after sitting for 4 long hours. 😖 I had to quick read or skip reading some answers or questions sometimes in order to make it within the allotted time.

It’s not only a test of your deep knowledge of AI systems and pipelines, but also of your stamina, perseverance, and endurance. Just like with my previous cert exams, I went to a test center, and I would highly recommend you to do the same if possible. Being able to have quick toilet breaks and drinking some water was of paramount importance to me, I can’t imagine staring at a screen for 4 hours straight.

Photo by JESHOOTS.COM on Unsplash

As to the covered topics, make sure you know RAG, Bedrock Knowledge Bases, and Bedrock Guardrails inside and out. There are tons of questions on those topics (and rightfully so: plenty of AI systems and projects involve these). Knowing the best logging/auditing solution also came up a ton (so make sure you know how to log metrics, API calls, FM reasoning…). Surprisingly, Agentic AI featured only in a minor way. That’s a bit sad, since Strands Agents and especially Bedrock AgentCore are pretty cool products if you ask me! This might change in the future, given that these two products are still relatively recent. Quite a few questions had Amazon Kendra among the possible answers, and a few had Amazon Lex. It was a bit of an indication that the answer that featured them was incorrect. 😁 These two services have become much more niche compared to modern AI services.

I honestly left the exam feeling like it could go either way. Time was tight and I often had to guess between what felt like two very close answers. What an outburst of joy I felt when I got the Credly badge email right before going to sleep!! 🤩 The next day I checked the results and my actual score was 760/1000, just 10 points above the passing score of 750! Phew! 😅 I would have liked to prepare a bit better, but prep time was tight during Christmas vacation, so it was either right after or weeks later.

Study materials I used

The most important resource in my preparation was undoubtedly Frank Kane’s course in Udemy. I had already taken Kane’s course for MLA-C01 and had felt comfortable with his teaching style. He’s been the first well-known 3rd party instruction to publish a course for this cert. It’s a bit unfortunate that the course consists of recycled materials from previous certs plus new lectures, without clearly distinguishing between the two. The rule of thumb is: if you can see Frank Kane’s face on the top right of the screen during the lecture, it’s most likely that it’s a new lecture made for this exam. Otherwise it’s a rehash of another cert (take those lectures only if you don’t know about the topic already, but if you have MLA and SAA fresh in your mind, you probably don’t need to). I took notes and made flashcards from these new lectures until I started to run out of prep time and had to rush past the remaining lectures, but I do intend on fleshing them out during the upcoming weeks.

The other real alternative to Kane’s course at this stage was the official AWS Skill Builder exam prep course, but even though I gave it a try, I just couldn’t stand its interface nor its AI-generated explanations and videos (if you could even consider its text “explanations”, it was mostly “make sure you understand how X service is used for Y”). I will say that the AI system scenarios presented in that course were quite interesting, and that the hands-on labs and Simulearn probably can teach you a lot. So maybe it’s not a bad idea to go through that course once you have learned the theory from Frank Kane.

I barely had one whole day for practice exams before the real exam (I recommend you reserve more time than I did!). So I was only able to do 30 questions from Tutorial’s Dojo first exam, and I did them in review mode (to immediately take notes). For some reason, TD questions featured Amazon Comprehend a lot, but it didn’t feature quite as heavily in the actual exam. I also did the 20 practice questions from AWS Skill Builder (the ones that are available for free). Even though I didn’t do a full practice exam, just this small exposure was really important for training. It also of course helps that I’m quite versed with AWS cert exams at this point.

Wrapping Up

Achieving this brand new AWS certification has been no easy feat, especially with the allotted prep time and the allotted time during the actual exam. But I truly enjoyed refining my AI skills and going deeper into RAG, vector stores, and agentic AI. Learning about responsible and secure deployments at scale also felt very relevant for real-world scenarios. This cert complemented well my real-world project experience in AI projects. And passing the exam was a great dopamine hit and a boost of self-confidence! 😁

I am releasing the Anki flashcards I already made for this cert as an early product in my Ko-Fi shop: https://ko-fi.com/s/22e0104816 They will initially be at a discounted price. I’m aiming to refine and polish my study notes and add new flashcards over the coming weeks, at which point I will release Notion notes, PDF notes, and comprehensive Anki flashcards for this cert. This will meet the standards of the study materials I already have in offering for my previous 4 AWS certifications: AIF-C01, CLF-C02, MLA-C01, and SAA-C03. Like I said before, I heavily encourage anyone to take these cert exams before attempting the Professional-level AIP-C01, they will be good milestones, plus you need a good foundation of AI/ML and AWS anyway. Use my materials if they can be useful to you, all available in my website https://christiangreciano.com .

If you will be attempting this cert, best of luck! 🍀

My Wishlist for 2026

2025-12-29T00:00:00Z

Photo by BoliviaInteligente on Unsplash

I’m writing this blog post on the last weekend of 2025. I will keep it short, but I definitely wanted to write before finishing the year, at least to wish everyone a happy new year 2026! And while we’re on topic of wishing, here’s a short wish list for 2026:

Let’s see AI finally create some new jobs. And I’m referring to something more than the joking “I can fix your vibe coded app” descriptions.
Let’s have some of those new jobs be ideal for juniors in tech who just need to get a foot on the door and gain experience.
Let’s see some platforms and algorithms promoting genuine human content instead of AI slop.
Let’s answer negativity with more positivity. No need to lie about anything, but we humans can choose to put a positive spin to pretty much any scenario.
Let’s get out of our comfort zone more often, especially when we know it’s the right thing to do. Don’t let algorithms and consumerism dictate all your choices, just because it’s the easy thing to do.

2025 has gone by really fast. I will always remember this year as the year I started to post seriously on my online blog and on my YouTube channel. But more importantly I hope to always remember that human connections are the most valuable asset any professional can have. As someone who has been fascinated with tech since I was a toddler, this is a truth I often need to remind myself about. Happy new year, everyone, and may your righteous wishes come true!

Avoid Free-plan AWS accounts, even if you're a beginner to cloud

2025-12-15T00:00:00Z

Although in 2025 AWS introduced the concept of Free-plan AWS accounts for new customers, in this blog post I want to make a case for using Paid-plan AWS accounts even if you are a beginner. This might be a controversial take since tons of people have asked AWS for years to implement a free playground/sandbox environment to learn AWS, and yet once they finally did, here I am now to tell you to jump straight to paid accounts. 😅

Free-plan AWS accounts

Historically, there were no Free- or Paid-plan AWS accounts. There were free tiers in AWS services, some of which had free usage for 12 months. Since 2025, AWS offers Free-plan accounts to brand new AWS customers: https://aws.amazon.com/free/ . In essence, Free-plan accounts won’t bill you during the first 6 months, and you get up to $200 of usage in AWS credits. However, once the 6 months or your credit usage is up, your workload will stop until you upgrade to the paid plan! Additionally, not all AWS services and features are available in Free-plan accounts.

Only ONE Free-plan AWS account per new customer!

Although you can have as many Paid-plan AWS accounts as you wish, a new customer can only use one Free-plan account once, and never again! In order to enforce this, AWS thoroughly checks your customer credentials (name, credit card, your billing address, your email…) and makes sure they haven’t been used in any other AWS account before, otherwise you won’t be eligible to open a Free-plan AWS account. Even if your previously used AWS account is inactive or has been closed long ago, AWS still keeps historical data of those accounts and this can prevent you from opening a Free-plan AWS account. Notoriously, you can use sub-addressing or mail extensions for creating different Paid-plan accounts, but NOT for creating different Free-plan accounts. For example, user+awsaccount1@gmail and user+awsaccount2@gmail both direct to user@gmail.com, and you can create two different Paid-plan accounts with those two email addresses, but you can’t create two different Free-plan accounts.

AWS Free vs Paid

Not all AWS services and features are available in Free-plan accounts, and your workload will be stopped if you exceed your credits or the 6-month limit. But at least you get the benefits of free usage, right? Wrong! Paid-plan accounts can ALSO benefit from up to $200 in credits, as well as the free usage of select services. Some AWS services offer free trials, others a free usage quota every month, and some are free indefinitely. And this is the same REGARDLESS of whether you use a Free-plan or a Paid-plan account. The only difference is that if you exceed your quotas, AWS will simply stop your workloads, instead of billing you while they continue to run. And if it’s about privacy, you still need to give your credit card details if opening a Free-plan account, even if you don’t pay anything! So, I don’t think the advantages are really there!

Learning AWS costs

Monitoring costs of your cloud resources, setting alerts, and managing budgets are essential skills for a cloud engineer. The sooner you learn them, the better. Any high quality online course or blog for learning AWS/cloud should teach you the good habit of setting budgets and alerts, as well as the good habit of tearing down your infrastructure the moment you don’t need it anymore. If the course you’re following did not teach you this in the beginning, you know it’s NOT of good quality! Letting AWS or other platforms create the guardrails for you is giving you the fish today instead of teaching you how to fish.

Be responsible! You should be OK with the idea of spending a bit of money learning AWS. You are, after all, running resources on someone else’s hardware. That’s NOT free! Learn to be mindful of your infrastructure: spinning resources up and then forgetting about them is a terrible habit. Most of us have had nasty bill shocks from stuff we didn’t know was billing us. You’d rather that be a couple of bucks while learning AWS than thousands of dollars in your company’s production environment! Choose to learn the hard lessons earlier rather than later. Googling “EC2 billing” or “S3 pricing”, reading the docs, and understanding how you will get billed should become second nature to you if you intend to use AWS/cloud seriously. You should also learn how using multiple accounts over a single monolithic account to host a complex cloud system is usually the way to go (can’t learn that in a single Free-plan account).

Bikes with training wheels are detrimental for learning

As an analogy to all that I’m advocating for, think about bikes with training wheels versus plain balance bikes. I used bikes with training wheels as a child because balance bikes scared me and I just wanted to enjoy riding my bike. I did not understand that I was learning bad habits and stunting my progress. I did not ride much as a child or as a teenager, and I did not venture into normal bikes until I was fourteen years old! And yes, even at that age, I still had to fall off from normal bikes a few times until I learned how to keep my balance. It would have certainly been better for me to learn to ride a bike properly when I was a young child!

When might a Free-plan AWS account make sense?

I can personally recommend a Free-plan account in a niche scenario: if you are a student with plenty of free time for the next 6 months, and you want to explore AWS to get a feel for it, then maybe you could go for a Free-plan account. I assume you’re not too serious about learning cloud, and you can dedicate quite a bit of time to play here and there during those 6 months. However, if you’re for example aiming to get certified in AWS, even if it’s just the Cloud Practitioner certification, I definitely will advocate that you train on a Paid-plan account. You won’t feel the 6-month time limit pressure, you will learn best practices on billing, and overall just be better prepared for the real world.

Conclusion

There are not many reasons why you should delay learning to monitor costs and budgets if you’re serious about learning AWS. Plus, tons of people learned AWS before Free-plan accounts started to exist in 2025. You certainly don’t need a Free-plan account to learn AWS, and it can even be detrimental to use one. Learning good habits and foundations in the beginning will save you lots of trouble later down the line.

The AWS GenAI Dev Pro (AIP-C01) Certification Beta is now live

2025-11-29T00:00:00Z

Last month I wrote a blog post with my initial impressions about the new AWS GenAI Developer Professional certification. We didn’t have much information back then, but we do now, and the registration for this certification’s Beta exam is now open. The earliest day you can take the exam appears to be 16th December.

I have read the official exam guide, started the AWS Skill Builder course, and taken the sample practice exam questions by Tutorials Dojo. In this blog post I will share what I know and have experienced so far.

Note on AWS Cert abbreviations

To avoid being overly verbose, I will use the following abbreviations for the AWS certifications mentioned in this blog post:

AWS Certified AI Practitioner → AIF
AWS Certified Machine Learning Engineer — Associate → MLA
AWS Certified Generative AI Developer — Professional → AIP

Certification Code: AIP-C01

First off, the certification code is now official: AIP-C01. Last time I mentioned how the cert’s name was a bit misleading, and now its accompanying code is also slightly confusing. “AIP” could be easily confused with the foundational-level “AI Practitioner” certification (with code AIF-C01). And since “AIP” most likely stands for “AI Professional”, why didn’t they name the certification “AWS AI Engineer Professional” in the first place? The chain from Foundational to Professional would also make more sense if it were at least AIF→AIA→AIP, but it’s now AIF→MLA→AIP… Yeah, no matter how you look at it, it’s far from consistent! 😅

Overview of Exam Guide

If you’ve taken AWS certification exams before, there’s nothing surprising in the AIP exam guide. It will all sound very familiar. It’s still mostly a multiple-choice questions exam with scaled scoring. Because the exam is currently in Beta and not General Access, it costs 50% less (i.e. $150), but it’s also longer (85 questions in 205 minutes) and you can’t retake it if you fail (you must wait until General Access to retake a failed Beta exam).

If you are going for the Early Adopter badge, this time it will be different compared to the AIF and MLA certifications: only the first 5000 people who pass the exam will be granted this Early Adopter badge. As I said in my previous post, the Early Adopter badge hasn’t brought me more value than just bragging rights, but if you’re going for it, be mindful that it’s a race with our peers. Is 5000 people a few or a lot? I honestly have no idea… 🤷‍♂

Maybe the most important part to take away from the exam guide are the 5 domains included in the exam, which you should cover well to increase your chances:

Foundation Model Integration, Data Management, and Compliance (31% of scored content)
Implementation and Integration (26% of scored content)
AI Safety, Security, and Governance (20% of scored content)
Operational Efficiency and Optimization for GenAI Applications (12% of scored content)
Testing, Validation, and Troubleshooting (11% of scored content)

AWS Skill Builder Prep Course

Third-party courses obviously don’t exist yet, but what we do have is the official exam prep course in AWS Skill Builder. We can expect content creators and third-party instructors to put out their materials in the upcoming weeks, especially after the first exam takers report on how the exam went, but in the meantime Skill Builder is the only trustworthy resource. I had never used Skill Builder before, but since I wanted to get a head start, I went ahead and paid a $29 one-month subscription and looked at the contents of the course. So far I have completed all the materials in Domain 1, and I look forward to completing Domains 2-5 in the next few weeks.

As someone who has prepared for previous AWS certs with third-party instructors and courses, I must say that the AWS Skill Builder UI is kinda garbage (similar to the AWS Management Console UI 🙄). There’s a ton of buttons to click to proceed, and you must enroll in the different domains (you can’t enroll in the whole course for some reason). In this screenshot for example you first must enroll in the course, then click “Start”, or else you’re just seeing a non-clickable table of contents.

Then there’s paying for a subscription to AWS Skill Builder, which gets billed to an AWS account. AWS accounts are for infrastructure and cloud resources, not for online courses, I don’t know why that should get billed there… Anyway, just saying that while edtech platforms like Udemy are not perfect, their UI is much more intuitive than AWS Skill Builder.

As to the content of the course itself, I do find it quite interesting! Each section has an example of a business scenario that requires a complex AI system, and every choice of AWS service, product, or feature used to build the AI system is outlined. They mix in checkpoint questions and videos to not have you just read endless text, which is appreciated. The videos are sadly AI-generated. At least the AI videos are of high quality, but I think it’s sad that a company like Amazon decides not to have a human instructor teach us, I’m pretty sure they can afford the salary of a well-qualified instructor! A great thing about Skill Builder is that it has some nice hands-on sections. While they do provide you with sample Python code and CLI scripts to set up the project, the labs don’t hold your hands completely, and the student still needs to take ownership of what they build and deploy. You can even post on LinkedIn with the hashtag #awsexamprep to have AWS personnel review your work and endorse your for the skills you’ve shown. If you have a paid subscription, you also get access to practice exams, flashcards (of mid quality, they look AI-generated to me), and AWS SimuLearn. SimuLearn are hands-on lab on steroids (powered by AI). I haven’t tried it yet myself, but I think I will want to give it a go, at least for one of the 5 domains (maybe if I feel I need reinforcement in a particular topic or area).

A word of warning: in order to make the most out of Skill Builder’s AIP course, I strongly suggest you take the preceding MLA-C01 certification! The AIP course assumes you have a strong knowledge of AI, AWS, and AWS AI services. It tells you how different AWS services, products, and features integrate together to build complex AI systems, but they don’t explain any service in isolation. Most of the course will fly over your head if you don’t know about those services, products, or features in the first place! Imagine reading about heart surgeries and lacking knowledge about veins, arteries, blood pressure, surgery tools, etc!

Image from Reddit

3rd-Party Content Creators have announced Courses

Good news! Some popular third-party content creators have announced that they are already preparing courses to help others pass the AIP exams. So if you prefer that way instead of AWS Skill Builder, you can wait a couple of weeks and there should be more resources out. Here are some examples:

Andrew Brown
Frank Kane
Tutorials Dojo
- TD also released a FREE sampler with 30 practice questions before the official exam guide was out. I went ahead and recorded myself answering TD’s sampler, so if you’d like to tackle the questions along with me, do check my video below!

I will attempt the Beta exam myself!

Since I did well in Tutorial’s Dojo sampler and I’m finding the AWS Skill Builder course interesting and fascinating, at this point I have warmed up to the idea of taking the Beta exam of the AIP certification. I will probably not take it on the 16th of December since I will already be enjoying some Christmas vacation, but I do think tackling it sometime in January is realistic. By then I hope that Tutorials Dojo has refined their practice exams and I can hone my claws and feel confident about the exam. It will be my first professional AWS certification, so in a way it’s also exciting (even if I’m not looking forward to a 4-hour exam! 😨). Stay tuned in my blog and social media to know how it goes for me in the next few weeks! And if you will attempt the exam yourself, best of luck! 🍀

I played LinkedIn Queens during a whole year

2025-11-10T00:00:00Z

For the past 365 days, I have completed the daily puzzle offered in LinkedIn Queens. No exceptions, although I do have two “freezes” if I would ever skip a day. It’s not a great accomplishment and it barely qualifies as bragging, but it does give me an excuse to write this blog post. I posted on LinkedIn a few months back when Queens itself turned one year old. And now it’s my uninterrupted streak that turns one year old. So let’s talk a bit about games and professional careers!

Games matter in professional settings

I have been a gamer my whole life. I vaguely remember picking up a NES controller at age 2 and enjoying making Super Mario jump around according to my button smashing. In particular, I’m a sucker for strategy games, whether in physical or digital form. I will highlight, in no particular order, the following games that are close and dear to my heart: StarCraft, Pokémon, Magic: the Gathering, Heroes of Might and Magic, and (plain old) chess.

I admit that these games and others have often been a time sink during my life, and game addiction is something to be mindful of. But at the same time, these games shaped my preferences for problem solving, improving via competition, and even tech. I was into programming since before I learned how to code. By creating custom maps in StarCraft and modding triggers and configurations, I was unconsciously learning to code and script. Years later, when I learned to code in Java, the programming patterns resonated strongly within me. Many others have similarly reported that games got them into programming, or they learned to program by creating games, or both.

Short games like LinkedIn Queens have many benefits

Games can often be temptations, distractions, and procrastination black holes when there’s more important stuff to do. You don’t need to ask me whether I would prefer playing a Magic: the Gathering booster draft or debugging a software app with a stack trace that’s longer than the Constitution. And that’s why short games like LinkedIn Queens can actually hit a sweet spot:

Rarely does a Queens session take me more than 5 minutes to complete
There is only one puzzle per day, so once you complete one, you can’t start the next until the next day (contrary to addictive online games which never end and try to keep you in at all costs)
It scratches my itch to solve a puzzle and awakens my mind to solve other problems
By checking the Leaderboard I see who else is playing the game, and I have even started conversations with connections I hadn’t talked to in years with this silly game as an excuse
It’s fun to share your results with friends and family (I share them daily with my wife and a Discord server of similar-minded nerds)
It forces me to open LinkedIn daily. This honestly is more in LinkedIn’s benefit than mine, but I do have all LinkedIn notifications silenced and no LinkedIn email reaches me, so it’s good to log in every day because that way I won’t miss any DMs by recruiters or connections.

I could go on and on, but short games have existed for a long time and the benefits are quite well-known at this point. Traditionally we got crosswords, word searches, chess puzzles, etc in physical newspapers. And in recent times we’ve had games like Wordle from New York Times. LinkedIn has jumped into this trend recently, and they’re continuing to add games, so they’re probably having the desired effect on their platform.

Be mindful of how much time you spend with games

Honestly, I’m not the best example for this counsel, but DO be careful with not letting games distract you too much from your important work. Just like with salt, a sprinkle of it on your food will make it taste so much better. But put too much salt and the food will be ruined (as well as your health!). LinkedIn has lots of games at its disposal but I currently only play Queens and Mini-Sudoku. I am usually done with both of them in under 5 minutes, so I know they won’t distract me too much. I used to play Wordle before, but I stopped because sometimes it could take me up to 15 minutes to figure out the solution (I just don’t like to give up or cheat…). It’s a shame because I do like word puzzles and language games.

What is your pet puzzle game you play daily? Reach out to me on social media and let’s talk about it!