The Algorithm’s Apprentice

They say you shouldn’t stare too long into the abyss because it might stare back. What they don’t tell you is that, in 2025, the abyss also wants to optimize your engagement, track your scroll rate, and gently nudge you toward buying another bamboo toothbrush.

My name’s Arman, and for the better part of a year, I worked as what my contract optimistically called an “Algorithmic Content Curator.” My friends called me “The Algorithm’s Apprentice,” and the nickname stuck—partly because it was funny, and partly because it felt painfully true.

I didn’t invent the algorithm, obviously. That was the work of engineers, data scientists, and the ghost of every tech visionary who once promised the internet would set us free. My job was simpler, yet stranger: to help the algorithm learn what people want, sometimes before they knew it themselves.

Each morning, I’d log into my dashboard, sip overpriced oat‑milk coffee, and be greeted by a swirl of data points: engagement scores, watch‑through rates, heat maps of where thumbs paused on videos. These weren’t just numbers. They were breadcrumbs through the forest of human attention.

It wasn’t glamorous. There were no hoodies and beanbag brainstorming sessions. Just me, my screen, and thousands of micro‑decisions: Should this video of a dancing cat get a boost? Should we downrank this clip where the language might offend advertisers? Should we recommend more DIY soap‑making videos to this user in Idaho?

At first, it felt harmless, even creative. Like helping the world discover things they might actually love. When we gave a tiny channel its first viral hit, I cheered. When a local artist’s song exploded because of a perfectly timed push, I felt oddly powerful, like a hidden hand moving the culture.

But over time, the edges blurred. I started to see people as clusters of probabilities: 82% chance they’d click on nostalgia content, 67% likelihood they’d share a story about outrage, 91% chance they’d ignore anything over three minutes. The algorithm didn’t care about nuance, context, or ethics. It cared about holding eyes on screens.

The hardest part was realizing I wasn’t immune. The same code I served was quietly shaping me, too. One morning, I caught myself checking engagement metrics before brushing my teeth. I started thinking in “content buckets”: heartbreak, humor, rage, hope. My own ideas felt like drafts waiting for an algorithm’s approval.

It came to a head during what I call the Great Outrage Spike.

One trending story—something silly, about a celebrity meltdown—started gaining traction. The metrics screamed: Boost it. Feed the beast. But I hesitated. The story was half rumor, half meme, and fully unnecessary. I clicked “pause,” a tiny act of rebellion.

My supervisor pinged me a minute later. “Engagement down 3% on that thread. Any reason?”

I typed, deleted, typed again: “Just thought maybe we shouldn’t amplify it.”

The reply was a single emoji: 🤷‍♂️

And that was it. The moment taught me two things: first, the machine doesn’t stop just because you do. Second, the apprentice can step away, but the algorithm will keep running—fed by billions of other clicks.

I left the job a month later. Officially, I said I wanted to “pursue creative projects.” Unofficially, I’d realized I was better at serving a machine than understanding myself, and that scared me more than unemployment.

Now, on the outside, I see the world a little differently. Every viral dance, every moral panic, every comforting nostalgia clip—it all feels less random. The algorithm isn’t some evil mastermind; it’s a mirror of us. But it’s a funhouse mirror: it stretches what we stare at longest, it shrinks what we neglect, and it sells tickets to watch ourselves distort.

Sometimes, friends ask what it was really like behind the curtain. I tell them the truth: it’s a mix of math and gut instinct, creative guesswork and raw data. And it’s addicting—because it feels like control, even if the control is mostly an illusion.

Would I do it again? Maybe. But this time, I’d remember: the apprentice serves the algorithm, but the algorithm is built from our choices. Each click, pause, and share becomes a vote for the world we want to see.

And maybe that’s the weirdest lesson of all: the abyss doesn’t just stare back—it scrolls with us.