Last week, I revised one of Unified.to’s earliest articles, An Overview of Unified APIs, rewriting it as What is a Unified API?
In addition to updating the text of the article, I also created some explainer graphics to liven it up and save the reader from being hit with just a wall of text. Those graphics are what you see in this article — enjoy!
If you were to time travel and visit Crazy Go Nuts University during my student days, you’d find that the thing I was known for wasn’t programming or playing the accordion, but drawing comics.
The web came around at the very end of my long and colorful academic career, so my comics mostly appeared in student newspapers — primarily Golden Words, a satire newspaper in the same vein as the original print version of The Onion, as well as the main student newspaper, The Queen’s Journal.
I make the occasional comic every now and again these days, and when Dan Arias, a former coworker at Auth0, found out about it, he asked me to draw some comics as a way to “storyboard” some screens for an app for the 2023 Oktane conference.
The comics were supposed to showcase some features of Auth0’s customer identity management system, and if possible, do so in a humorous way. They also had to use some animal mascots that had been created for the project: a platypus, a rabbit, a capybara, and a boar.
I recently found the sketchbook with the comics I made for the app. They never went into the app — they were just storyboards for the app’s artist, Sofía Prósper Díaz-Mor, to use as guides, and the final versions that appeared in the app looked fantastic.
Still, there’s a rough charm to my doodles, so I thought I’d post them here. Perhaps it’s time for me to make more posts as comics…
The app had a space theme, so all the comics featured our animal characters — once again, a platypus, a rabbit, a capybara, and a boar — as characters having science fiction adventures that also featured some aspect of digital identity.
This comic was about fine-grained authorization, which is a fancy way of saying “very specific control over who’s allowed to do what in a system”…
This comic was the storyboard for a story about anomaly detection, which attempts to detect logins that have a suspicious quality to them. I did this by having an alien disguise themself as the ship’s commanding officer, Captain Platypus, and board the ship…
“The planet of a thousand apps” was the setting for this comic about single sign-on. The idea was every activity on the planet was controlled by its own app, which meant that you either had to log into a different app to do anything, or you could use single sign-on…
To illustrate the security advantages of passkeys, I came up with this comic. It shows that with a passkey, you don’t have to memorize a password, and even if a hacker manages to break into the server, all it has is the passkey’s public key, which (as its name implies) is known to everyone…
“Make Star Wars without getting us into legal trouble,” they said, and this is the resulting comic. It features our rabbit character as “Bun Solo” and our capybara as “Capybacca.” In this rough sketch comic, they destroy the centralized identity database, the Data Star, freeing the citizens of the galaxy to use decentralized identities. In the second page, I show the uses for them…
Watch this space — I think it’s time to do more comic-style blog posts here on Global Nerdy!
If you’ve tried to go past the APIs like the ones OpenAI offers and learn how they work “under the hood” by trying to build your own neural network, you might find yourself hitting a wall when the material opens with equations like this:
How can you learn how neural networks — or more accurately, artificial neural networks — do what they do without a degree in math, computer science, or engineering?
There are a couple of ways:
Along the way, both I (in this blog) and Tariq (in his book) will trick you into learning a little science, a little math, and a little Python programming. In the end, you’ll understand the diagram above!
One more thing: if you prefer your learning via video…
My third experiment for 2024 involves trying out the ideas from Noah Kagan’s new book, Million Dollar Weekend.
ℹ️ In case you’re wondering: my first experiment of 2024 was to turn my layoff experience into a series of articles; the second was to take a chance working with a pre-seed startup.
Why conduct such an experiment? For now, let’s just say that current circumstances make it necessary, and hey, if anyone can pull off this kind of thing, it would be me.
The general idea of Million Dollar Weekend is that you can start a lucrative business by doing the following:
The prerequisite for the Million Dollar Weekend process is a certain amount of unmitigated gall. Time and again in the book, Kagan states that two things hold people back from starting businesses:
Kagan’s methodology is to start by trying out an idea, seeing if someone will pay for that idea, and then either refining that idea or coming up with a new one and repeating the cycle.
The methodology anticipates rejection, and in fact, it says that in selling your idea, you should aim for plenty of rejections. The idea is that if you’re getting rejected often, you’re asking often, and that’s what eventually leads to success.
I’ll write more as I continue with this experiment, but for now, if you’re curious, here are some resources I can point you to:
You might also find these interviews with Kagan interesting:
ℹ️ Also in case you were wondering: This is NOT a paid promo for the book — neither Noah Kagan nor his businesses have any idea who I am or how to deposit money into my bank account. I wish they did!
Masterminds Tampa Bay is holding their AI Superpowers Unlocked panel on Wednesday, May 15th with the following panelists:
Here’s Tampa Bay Masterminds’ description of the event:
Unlock the future of Artificial Intelligence at “AI Superpowers Unlocked: An Expert Panel,” an event meticulously crafted for entrepreneurs, tech enthusiasts, and forward-thinkers ready to explore AI’s transformative potential. Join us to gain practical insights on becoming a leader in AI application and connect with industry pioneers.
AGENDA
- 6:00 PM – 6:30 PM: Socializing Time
- 6:30 PM – 7:15 PM: Expert Panel Discussion
- 7:15 PM – 8:00 PM: Audience Q&A / Conclusion
Main Takeaways:
🧠 Understand the crucial role AI plays and why mastering it is essential.
🧠 Learn strategies to best leverage AI for 2024 and beyond.
🧠 Discover essential AI tools beyond ChatGPT.
🧠 Explore best practices, ethics, and more through interactive FAQs.
Why You Should Attend:
🚀 Tailored for Forward-Thinkers: Designed for those poised to disrupt markets and lead innovations, this panel will help you stay ahead in the AI curve.
🚀 Unparalleled Insights: Spend an hour with AI luminaries discussing strategies and visionary applications to outpace competitors and drive success.
🚀 Networking Opportunity: Connect with like-minded professionals and innovators, and perhaps discover your next great collaboration.
This is a paid event — attendance is $35 and supports Tampa Bay Masterminds’ mission of fostering innovation and education in technology, with all ticket sales considered donations.
Want to attend? Register at lu.ma/superpowers!
It’s been five years and a couple of jobs since I’ve had business cards, but as Unified.to’s Supreme Developer Advocate, I’ve been issued a set, and they arrived in the mail earlier this week.
They came in nice packaging…
…and rather than the traditional rectangular design, ours are square. On the front is Unified.to’s octopus logo and mascot…
…and on the back in Unified.to’s QR code and URL:
And, damn, do I like that box. It reminds me of this tweet about boxes and being a grown-up:
I don’t know much about Jukebox, the company that made these cards, but it seems that they’ve matched MOO (whose cards I’m familiar with from a handful of previous companies) and beaten them at the unboxing experience.
By popular demand, here are the “slides” from my presentation this morning at Civo Navigate Local Tampa, Make Smarter AI Apps with RAG!
Retrieval-Augmented Generation, also known as RAG for short, is an AI technique that combines…
…to enhance or improve the responses generated by the model.
At this point, you’re probably thinking this:
This talk runs from 11:15 to 11:30 a.m., which is just before lunch, and I’m not at my cognitive best. Can you explain RAG in an easy-to-digest way, possibly using Star Wars characters?
I’m only too happy to oblige!
Consider the case where you ask an LLM a question that it doesn’t “know” the answer for. The exchange ends up something like this:
With retrieval-augmented generation, you improve the response by augmenting the prompt you send to the LLM with data or computation from an external source:
Because RAG provides additional information to the LLM, it solves two key problems:
Here’s a lower-level view of RAG — it starts with the cleaning and conversion of the supplementary data:
Once that supplemetary data has been cleaned and converted, the next step is to convert it into small chunks of equal size:
Those chunks are then converted into vectors. If you’re not really into math but into programming, think of vectors as arrays of numbers. Each of the numbers in the vector is a value between 0.0 and 1.0, and each vector typically has hundreds of elements. In a diagram below, I’ve greatly simplified the vectors so that they’re made up of only three elements:
The whole process of cleaning/converting, then chunking, then embedding is called indexing:
Now that you know what’s happening “under the hood,” let revisit the RAG diagram, but with more detail:
Here’s what’s happening:
That was the “hand-wavey” part of my lightning talk. The rest of the talk was demonstrating a simple RAG system written in Python and running in a Jupyter Notebook. If you’re really curious and want to see the code, you can download the Jupyter Notebook here.