Winning Your Kid's Heart and Sparking Imagination with GenAI
An interview with AWS Hero Agustinus Nalwan on how he used Amazon Bedrock to create a comic video generator for his son.
Published Jan 12, 2024
AWS Hero Agustinus Nalwan (Gus) treasures spending time with his son, and would do anything to delight and surprise him. As a busy parent and an Artificial Intelligence and Machine Learning expert, he saw an opportunity to solve problems right at home. How could he keep his son's imagination and curiosity alive through stories every night and still manage his other priorities? He did it by building an AI comic video generator with Amazon Bedrock! The story is so heart warming and the innovation so compelling that it became the subject of a talk he gave at AWS re:Invent 2023. Below, find the recording of Gus' talk complete with how you can build something similar, and a brief interview delving into why this was important to him.
Who can learn from your talk and what is the most important thing you want them to do differently after hearing it?
"Anyone who shows interest in tech can learn from my talk. I'd like to convince people that building AI tech does not require you to possess Phd degree nor a deep software engineering background. Framework and tools are getting easier by the day, using a few lines of code you can glue things together to get the job done. Especially using tools and services from AWS."
What are your thoughts on where builders should put their attention with GenAI, considering so much is happening in this area right now?
"The builder should focus on:
- Learning how to orchestrate application with LLM using techniques like prompt engineering, Retrieval Augmented Generation and Agent/Assistant.
- Building a robust evaluation pipeline as you will be conducting lots of experimentation with your prompt, parameters and techniques and you need to be able to measure your model performance efficiently.
- Gaining insights from others about possible applications. Gen AI opens up new avenues that were previously unattainable. It redefines the boundaries of what is achievable."
Is there anything someone should understand first before watching your talk, and are there community content resources that would help them prepare?
"Having some programming knowledge is likely necessary, especially for AI/ML, where Python is the preferred language. Therefore, blogs or tutorials that instruct on Python coding can be very beneficial. However, don't limit yourself to just reading or watching these tutorials. It's important to actively engage with the material. Apply what you learn by creating projects; this hands-on approach will enhance your understanding significantly."
You can check out some of the videos created by Gus' generator on his channel here or learn from him on his Medium blog.
What didn't you cover in the talk because time was limited that you wish you could have?
"If I had more time I would probably take you step by step through the code within the Jupyter notebook to fine tune the Stable Diffusion 2.1 model and also the code I run to orchestrate the whole thing using AWS Batch."
What's one question you wish someone would ask you about this topic?
"How long does it take you to build Owly tech? Surprisingly it is done within less than a week, assuming 8 hours a week, 5 days a week. Most of the fine-tuning code has been prepared for you from Amazon SageMaker JumpStart."
How did you become an expert in this area, and why is it an area you are passionate about?
"I am super passionate in solving problem using tech and I always like to automate as much as possible. This is the drive for me to learn more. You can automate lots of process with code but AI takes automation to the next level. It builds the algorithm for you from your training set. So obviously AI is an area I am currently passionate about."
Agustinus Nalwan is the GM of AI, Data Science and Data Platform at Carsales. He resides in Australia and lives by the motto, "Nothing is impossible. It is just not possible yet". Connect with Gus on LinkedIn or learn from him on his Medium blog.
AWS Heroes are a community of seasoned technology builders outside AWS who have contributed incredible efforts to their field, helped others learn and grow in communities and tested and provided critical feedback to AWS on its services. To connect with an AWS Hero in your area or in a particular field of expertise, visit the program page here. AWS User Group meetups are an easy way to start connecting with and learning from community. Find a meetup near you.
Any opinions in this post are those of the individual author and may not reflect the opinions of AWS.