Everyone is going to have a nonlinear career
what a viral AI post gets right vs. wrong, and what I'd do about it
Hi, I’m Hannah! Welcome to Nonlinear News, where I write for smart, ambitious people making big pivots.
💼 This week’s jobs in tech/startups (as always, reply to this email if you have a role you’d like to share with the community!)
Matt Shumer, an AI founder, posted an essay on X this week called “Something Big Is Happening” that now has more than 80 million views. I posted a video about it and a lot of you saw it, but short-form video doesn’t leave much room for nuance, and this topic needs it. So I want to dive deeper here.
Despite being a big vibe coder and AI dabbler in my own life, I don’t write about AI in this newsletter. It’s not my expertise and I already spend way too much time thinking about it in my 9-5. But I’m making an exception because I think a lot of you are in industries like finance and professional services where nobody is having this conversation. Frankly, even at most tech companies and startups, people aren’t having it either.
After reading the article, the counterpoints, and sitting with it for a few days, the takeaway I keep coming back to is this one:
Everyone is going to have a nonlinear career soon, not by choice, but because the linear path is going away.
“The honest version sounds like I’ve lost my mind”
Shumer said he wrote this article for his parents, not for tech Twitter. He said he keeps getting asked “what’s the deal with AI” and keeps giving the polite version, because “the honest version sounds like I’ve lost my mind.”
The honest version: he says he is “no longer needed for the actual technical work of my job.” He describes what he wants built in plain English, walks away for four hours, and comes back to the finished thing with nothing to correct.
He’s writing to people who tried ChatGPT a year ago and thought it wasn’t impressive. He says the free version is over a year behind what paying users have access to. And what he’s talking about isn’t the chatbot you ask simple questions. He’s talking about AI agents that control your browser, open multiple apps, and do your work for you.
He lays out this timeline:
“In 2022, AI couldn’t do basic arithmetic reliably. It would confidently tell you that 7 × 8 = 54.
By 2023, it could pass the bar exam.
By 2024, it could write working software and explain graduate-level science.
By late 2025, some of the best engineers in the world said they had handed over most of their coding work to AI.
On February 5th, 2026, new models arrived that made everything before them feel like a different era.”
He says this is happening in tech first but lists out every industry it’s coming for: law, finance, consulting, customer service, writing. Anything you do on a computer, and not in ten years — in one to five. He cites Dario Amodei, the CEO of Anthropic, predicting 50% of entry-level white-collar jobs gone within that window. And he says people in the industry think Amodei is being conservative.
The pushback
This shouldn’t be a one-sided conversation, and there were some responses this week that I think are worth reading.
John Coogan wrote a piece on TBPN called “AI Is Not Covid”, rejecting Shumer’s comparison of the AI revolution to Covid, and I think his core point is fair. Covid could infect anyone at any time, which is why it made sense to warn everyone. AI disruption won’t work like that. It won’t hit everyone at the same time or in the same way. A surgeon’s patients still want her holding the knife. A teacher isn’t getting replaced by a robot next school year.
Where I agree: the urgency is different depending on how close your work is to a screen. If you’re a junior analyst building models in Excel all day, your timeline is very different than a surgeon’s. If you’re in an operations role that’s mostly email and spreadsheets, you should be paying more attention than someone whose job requires them to physically be in a room with people.
Where I’d push back: I think Coogan underestimates how many people’s jobs are closer to a screen than they think. And the fact that disruption isn’t evenly distributed doesn’t mean you should wait to see if it hits you. By the time it’s obvious, the window to prepare has closed.
Then there’s Will Manidis, who wrote a piece called “Tool Shaped Objects” that I think is the most interesting counterpoint. His argument is that a lot of AI right now functions as what he calls a “tool-shaped object” — something that looks like a tool, feels like a tool, produces the sensation of work, but doesn’t actually produce work. He compares it to FarmVille: you click, the number goes up, you feel productive, but nothing was built. He says people are setting up elaborate AI workflows whose primary output is the existence of the workflow itself. The setup is the product.
I’ve seen that. But I think it actually strengthens the case for paying attention. If some people are going to waste time playing AI FarmVille while other people are using these tools to do real work in a fraction of the time, that gap is going to widen.
Five things I’d start doing
Despite the pushback, Shumer gives concrete advice in the article and I agree with most of it. Here’s what I’d tell you, combining his suggestions with my own.
Start using AI for real. If you’re on the free version of ChatGPT, you’re falling behind. It’s like how only your mom and grandma use Facebook now. Get Claude, download Claude Code, and learn how to use Claude Skills, which is how you give AI specific instructions to do real work for you. Use the latest model, not the default. Don’t ask it questions you can google — feed it a contract, a spreadsheet, an actual doc from your job. Go vibe code something with Lovable.
As Shumer writes: “if [something] even kind of works today, you can be almost certain that in six months it’ll do it near perfectly.” Spend one hour a day doing this — not as a search engine, but in your actual workflows. He says if you do this for six months, “you will understand what’s coming better than 99% of the people around you.”
Invest in relationships. The most valuable people in 2-3 years will be the ones who can connect pieces across teams and stakeholders, not the ones who can build the best spreadsheet. Trust built over years, connecting dots across people, reading a room — AI can’t do any of that. Roles like sales, chief of staff, customer success, parts of product management and marketing are still going to matter, even if they have different titles.
Get your finances in order. As Shumer puts it: “if you believe, even partially, that the next few years could bring real disruption to your industry, then basic financial resilience matters more than it did a year ago.” Build up savings, be careful about new debt that assumes your current income is a given, and give yourself options.
Start building something for yourself. I’m not saying quit your job tomorrow. But you have a window right now where you still have income and stability, and a year from now that might look different. Start a side project, build a small audience, create something.
Build a personal brand. If entry-level roles in finance, consulting, and law are disappearing, then “analyst at X firm” stops being an identity. Your job title is not going to carry you. Being known for something will.
Every career will be nonlinear
A lot of you follow me because you have nonlinear careers, or you want one. You’re in finance but thinking about tech. You’re in consulting but want to build something. You’ve changed jobs multiple times and people keep asking what you actually do.
This AI wave is going to make that the norm, regardless of how big you think the impact will be. The stable, linear career path most people spend their whole lives optimizing for is in the line of fire. Everyone will have a nonlinear career, even if they didn’t choose one, because the linear path stops existing.
But if you've been reading this newsletter for a while, you already know how to
operate in that world. The five things above are the same things that have always separated people who navigate change well from people who get caught off guard: use the tools, invest in people, stay financially flexible, build something of your own, and make sure you're known for more than your job title.
And above all of them is the one skill that always compounds: learning how to learn fast. The people who will be fine aren't the ones with the best current job or the most stable industry, but the ones who can pick up new tools, adapt to a new normal, and figure things out in real time.
Reads, Ideas, Tools
“Something big is happening” by Matt Shumer
Jobs & Resources
cool jobs for ambitious nonlinears (updated ~weekly)
This resonates deeply. The shift from "choosing" a nonlinear career to having one thrust upon you is probably the most important framing here.
💝