How I passed AWS AIP-C01 (Generative AI Developer Professional) in Beta
1414 words (~9 mins read)
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! 🍀
- ← Previous
My Wishlist for 2026