If you’re a non-engineer experimenting with prompt engineering, here’s the deal: understanding how LLMs actually work isn’t just a nice-to-have, it’s non-negotiable.
Take the Chain of Thought technique as an example. Using it after the model has already decided is like trying to screw in a bolt with a hammer, it’s just not going to work. At that point, the model isn’t “thinking” or “deliberating” anymore. It’s already locked in its answer and might just be scrambling to justify something wrong.
Sounds obvious, right? But tons of articles and tools (looking at you, Langchain) get this wrong. They miss the timing and end up using models to evaluate their own responses in ways that don’t make sense, leading to garbage conclusions.
If you want to build better AI solutions, get the basics down. Knowing how LLMs tick will save you headaches, lead to smarter results, and happier users.
Here’s a question: if tech is all about building the future, why does it still look like a reunion of the same dudes from your high school coding club?
Let’s be real, diversity isn’t just about being nice or ticking a box. It’s about making better stuff. When the same types of people build everything, we end up with products that don’t work for everyone, like AI that can’t recognize darker skin tones.
I came across some stories that really made me think about who’s not at the table and why that’s a problem:
Meet Jodi Kovitz – Jodi’s making sure women get a seat at the table (or just build their own table).
Meet Saadia Muzaffar – Saadia’s here to remind us that breaking barriers isn’t optional, it’s necessary.
These stories aren’t just inspiring; they’re a wake-up call. The more voices we bring into the room, the better we all do. So why are we still moving at the pace of molasses?
Let’s ask ourselves: who’s not here, and what are we losing because of it?
Living on an island farm wasn’t in the plan. It sounds like something out of an indie movie, but there I was, sharing space with a bunch of animals and a horse named Camila. A small horse, probably a pony (?). Her main job was keep the grass low and turn it into manure, that’s it. I was glad that people weren’t riding her. The one time I saw someone try, I hid her saddle and said it was stolen. First time admitting that. Hope her owner never reads this.
While she was getting her job done, I was coding and designing in my own unconventional way. No fancy degrees, just a lot of trial, error, and late-night debugging. Friends called me a “GoHorse” developer. They laughed. I laughed louder. It wasn’t a joke to me. It was an honest way to work, none of that “fake it till you make it” nonsense.
One Saturday, feeling burned out, I went outside to check on Camila. The farm was buzzing, chickens clucking, cows mooing, birds having a full-blown concert. A different kind of debugging.
In the quiet of the stable, Camila looked at me, unimpressed. I scratched behind her ears, grounding myself in something real.
Say Hi to Camila! 𐂃
I led her into the paddock, her brownish fur catching the sun. Her shining eyes, she had the most incredible black eye, deep and knowing. I loved that eye. As we walked, I thought about unicorns, the mythical distractions I’d chased before. Shiny big ideas, big promises, mostly letdowns. Horses, though? They’re the real deal. No sparkle, just steady work.
She stopped and stared at a distant hill. I followed her gaze, a wildflower pushing through the cracks of an old wall. It reminded me of half-baked ideas that somehow survive, despite the chaos.
“Maybe it’s not about chasing the impossible,” I thought. “Maybe it’s about nurturing what’s real.” Camila nickered. Either she read minds, or she was just hungry.
At one point, I tripped over a root, landing in the grass like a broken script. Camila gave me a nudge, as if saying, “Bugs happen.” I laughed. Laugh at myself is my second language, after JavaScript.
As the sun set, I realized something. Horses don’t promise the moon. They’re just there, solid and dependable, ready to move forward. Unicorns? They leave you empty-handed when reality kicks in.
I spent the last hour with Camila, untangling ropes, telling her about my latest project fails. No buzzwords, no pretense, just something real.
Walking home under the twilight sky, I felt clearer than I had all week. Maybe it’s time to build more horses in my life. Real, steady, built to last.
After all, sometimes the most magical things don’t need a fairy tale to prove their worth.
So here I am, back at my desk, inspired by Camila and the wildflower. I’m here for real horses, not unicorns. I’m here to break even, not break hearts.
Here’s what happened with an experiment I ran with the new council members in 2016.
In Florianópolis, each council member had a gross monthly budget of R$15,759.66/month to hire between 4 and 10 advisors.
To promote transparency and public participation, in early November 2016, I sent an email suggesting they adopt an open and transparent selection process. The goal was to avoid political favoritism, foster accountability, and give well-qualified individuals a fair chance at these positions.
But this wasn’t just any email. I included a “reading check” to track how many times the email was opened and who actually read it. The results? Some politicians opened it multiple times, while others didn’t even bother.I also tracked the response rate; let’s just say I’m still waiting for answers from most of them.
Do you still remember who you voted for? Why not ask them to publicly share who their advisors will be and what roles they’ll have? Transparency starts with us.
The sewer system is scheduled to be implemented in Ratones a neighborhood surrounding Florianópolis’ largest watershed by 2032.
Currently, most homes rely on traditional septic tanks or discharge directly into the water system, much like other neighborhoods in the city. It feels like we’ve skipped a few steps in our complex journey toward becoming a better society.
We bypassed the Middle Ages and now aim to be the Silicon Island. Yet, we have the worst water quality among Brazilian capitals and are 20 years behind when it comes to sewage and waste management.
However, I believe this could be an excellent opportunity to innovate and become a reference in sanitation and waste management technology. But in a culture that has never prioritized these issues, it would take more than 20 years for people to change spontaneously. That’s why I believe in public policies to transform our Medieval mindset.
This is the only topic that interests me in these elections. I spent this rainy Sunday analyzing candidates’ proposals and found people who share this concern, I found just one, whose platform includes the following goal:
“Advocating for decentralized and ecological sanitation systems with biological filters in small stations, as well as encouraging efficient household systems such as wetlands, banana circles, dry toilets, evapotranspiration basins, etc.”
The “etc.” here should really include “throwing the s**t at the fan” by holding the Medieval criminals still sitting in the Chamber, City Hall, Fatma, and Casan accountable those who have achieved only 15% of their annual sanitation goals for the past 20 years.
