Flip That Totem Pole

Started From the Bottom …

Breaking into programming is a special kind of hell. There are a million things to learn, and you will suck at most of them. Worse, you have no clue what the actual job is like, so you’re studying in the dark. I made two massive rookie mistakes.

Mistake one: I waited way too long to start applying. Every month I hesitated was another month of silently telling myself, you’re not good enough yet.

Mistake two: I didn’t think about what the market actually needed. I dove deep into Python/Django because it felt modern and cool. But where I lived–in the land of alligators and humidity (a.k.a. the Southeast)–employers were stuck on Java and C#. My stack was crickets in the local job market.

Applying to the same three job listings over and over wasn’t cutting it, so I got scrappy. While browsing Python’s official documentation (yes, I am that nerd), I stumbled onto a hidden gem: a job board. There it was, a Python-related gig in Tampa, FL. I emailed the HR director, attaching my résumé along with a very confident pitch about why I’d crush the job. A few days later, I had an interview.

Now, this was Tampa, not Silicon Valley, so I wore a suit. I sat down with the lead engineer and VP and braced myself for the interrogation: Python, Django, SQL, web protocols, REST APIs–you name it, they threw it at me. Luckily, I had a killer trump card: my experience running a real-time, at-scale business where my own money was on the line. They didn’t just hear skills–they saw stakes. I got an offer the next day.

That was my first programming job. And let me tell you, starting from the bottom feels a hell of a lot better when you get to stand on your own two feet.

Honeymoon Period

Sometimes you just get lucky. And man, I hit the jackpot with this job. It was at a traditional advertising company, but the dev team working on a new project had our own private slice of tech heaven. Imagine this: a massive corner office decked out in full Big Tech glory–bright red walls, electric standing desks, ergonomic chairs that whispered “productivity,” dual 4K monitors, MacBook Pros, a fridge stocked with snacks, a couch for power naps, and even a TV and games. It was like stepping into a startup oasis in the middle of corporate Florida.

The vibe was electric, and the team was sharp. I was the low man on the totem pole–not quite a junior dev, but definitely the new kid. That meant I got all the crap work nobody else wanted to touch. And you know what? I loved it.

See, everyone wanted to play with the shiny stuff–real software development, as they called it–not wrestle with Facebook’s busted-ass API. But the app we were building couldn’t run without ads, and this was an advertising company, after all. So guess who got tasked with the thing that actually makes money?

I’ll give you a hint: I like making money.

The problem was Facebook Ads' Learning Phase. Nobody could figure it out, mostly because Facebook’s Marketing API was a mess. The error messages were cryptic, the documentation was garbage, and the whole thing was glitchy as hell. But I had experience with what might still be the worst API in existence–eBay’s. (SOAP and REST at the same time? Why not just kick me in the shins?)

They gave me two months to crack the Facebook API. I figured it would take two weeks. It took two days.

When I showed my work, the team was stunned. The senior dev who was supposed to be mentoring me–let’s call him Chris–didn’t believe I’d done it. He tore through my code like he was searching for contraband, desperate to find some mistake or shortcut. But software doesn’t lie. If it works, it works.

Suddenly, I was the office’s little superstar. And anyone who’s worked in corporate America knows the rule: the reward for doing great work is… more work.

Fine by me. I’d spent years grinding seven days a week, working until I passed out. Now they were letting me leave at 5? And I got weekends off? This wasn’t work–it was a vacation.

If You See Something, Say Something

Once I’d automated Facebook’s API (and breezed through the much friendlier Google API), I was finally free to work on the SaaS app. Everyone was buzzing–this thing was the project, and we were all hyped to ship it as soon as possible.

That’s when I had my first run-in with a phenomenon I’d see again and again throughout my career: design by dictatorship.

The VP, a marketing guy with little tech experience, was calling the shots. We had a talented designer on staff–this was pre-Figma, but it wouldn’t have mattered because the designer wasn’t designing. They were translating every whim of the VP into something pixel-perfect and insanely over-engineered. Hours each day were spent obsessing over micro-components, tweaking screens, and making wholesale changes to layouts. It was like watching someone rearrange deck chairs on the Titanic–beautifully pointless.

Here’s the thing: it doesn’t take a genius to know how it looks isn’t the same as how it works. But with enough HTML and CSS magic, you can make it seem like it works.

Morale was sky-high. The whole office was running on the tech equivalent of cocaine, convinced we could launch in two weeks. “Let’s do this!” the VP proclaimed in our planning meeting. Then he laid out how much cash they were about to pour into marketing for the big debut.

I sat there, looking around the room, waiting for someone–anyone–to point out the obvious. We weren’t even close to being ready. Nothing was integrated, half the features were broken, and the other half only worked in carefully staged demos. But nobody said a word.

So I did.

“I don’t think the software will be ready in two weeks,” I said, breaking the buzz.

The room froze. Chris, naturally, told me I was wrong. Dead wrong. I’d just thrown a grenade into the meeting, and it exploded into a full-blown shitstorm.

Later that day, the owner of the company–let’s call him Anthony–called me into his office. I braced myself for the fallout.

“Why do you think the software won’t be ready?” he asked.

I laid it all out: the gaps, the bugs, the over-designed mess we were trying to duct tape together. “I’ll bust my ass to make it happen,” I told him, “but software doesn’t care about opinions. Either it works, or it doesn’t. We don’t need to argue–we’ll find out soon enough.”

Two weeks later, we found out. I was right.

Lesson learned: when you see a train about to derail, speak up–even if it makes everyone else hate you.

… Now I’m Here

One of the perks of a privately held company with a single owner and no investors? Decisions happen fast. Anthony didn’t waste time. He put me in charge of the team. Overnight, I went from the lowest-ranking developer to leading everyone except the VP.

As you can imagine, this didn’t go over smoothly.

The designer quit immediately. Literally the day before, we had a great working relationship, but she said she couldn’t see herself working for me. The VP–already fuming from my earlier call-out–completely lost it. He threw out a stream of insults, calling me “slick” and other not-so-subtle coded language. (It’s Florida, after all.)

The rest of the team? Mixed reactions. Some were professional, but others gave me the kind of side-eye you feel in your spine.

Then there was Chris. I’ll never forget how red his face got when he cornered me in the stairwell, screaming loud enough for the whole 200-person office to hear. He unloaded a lifetime’s worth of insecurity: You’re not shit, you don’t know anything, you’re not technical enough. I’d hear this kind of garbage the rest of my career, but he was the most explicit.

It took a while, but eventually, I fired him. Not for yelling at me, though. I could handle that. I even turned a blind eye to the Adderall he’d casually distribute to the team like party favors. But when the other drugs came out, that crossed a line even I couldn’t ignore.

That first engineering manager role was brutal. The team had 80% turnover. My boss–the VP–hated me openly and made sure everyone knew it, every single day. But that adversity forged something unique.

The team that stayed? We took on an us vs. them mindset. The famous underdog mentality you see in sports. Nobody believed in us–not the VP, not the designer, not even most of the company. But we were determined to prove them all wrong.

The codebase was a mess, far from the pixel-perfect designs we’d been handed, but it worked. And it did more than just “work.” We built a SaaS platform that could inject real-time data–things like flight prices or weather–into Facebook and Google ads and automatically adjust ad spend based on what the data said.

Years before Google Performance Max or Meta Advantage, we were running automated digital ad campaigns. At scale. We powered international campaigns for accounts as big as Home Depot and Saint Lucia.

It wasn’t glamorous, and it damn sure wasn’t easy, but we got the job done. That’s the thing about starting from the bottom and flipping the totem pole–you figure out what really matters: results.