How to stay relevant as a developer
<p>We are seeing a trend in layoffs in perfectly healthy companies. Blocks laid off nearly half their staff, and it’s not because they can’t pay them:</p> <blockquote> <p>Gross profit continues to grow… but something has changed. We're already seeing that the intelligence tools we're creating and using, paired with smaller and flatter teams, are enabling a new way of working which fundamentally changes what it means to build and run a company.</p> <p>— <a href="https://x.com/jack/status/2027129697092731343" rel="noopener noreferrer">Jack Dorsey</a></p> </blockquote> <p>We’ve all heard it: AI is coming for your job. This looks like Exhibit A, doesn’t it?</p> <p>There’s a real point buried in the hype. AI will happily eat the tedious, time-heavy parts of writing code. And if “I write code” i
We are seeing a trend in layoffs in perfectly healthy companies. Blocks laid off nearly half their staff, and it’s not because they can’t pay them:
Gross profit continues to grow… but something has changed. We're already seeing that the intelligence tools we're creating and using, paired with smaller and flatter teams, are enabling a new way of working which fundamentally changes what it means to build and run a company.
— Jack Dorsey
We’ve all heard it: AI is coming for your job. This looks like Exhibit A, doesn’t it?
There’s a real point buried in the hype. AI will happily eat the tedious, time-heavy parts of writing code. And if “I write code” is the entire value you bring, then AI is going to replace you.
But if you’d like to be better than that, here are a couple of ideas.
Ship faster
Instead of working less, put in the same effort and ship more. Teams will hit the same output with half the people, so you need to be twice as productive.
Efficiency doesn’t shrink demand; it expands it. The easier it is to build something, the more things your company will decide are "worth building." This is called Jevon's Paradox.
But don't ship AI slop
AI can write complex regex (so glad I never learned to do that) and can work with HTML canvas, but it's still a junior dev in many ways. Inefficient queries. Repetitive code.
LLMs lack context, even if you refine your claude.md. They often don't know that there's existing code they can reference to implement a similar functionality unless you've specified it. Maybe you want to structure your database schema for a feature you have in mind that isn't implemented yet.
Your job is to spec, review and keep the codebase quality high.
Expand your skillset, and collaborate less
Frontend devs can do backend now. Full stack devs can learn design. Instead of handing over your work for someone else to fill in the gaps, do it yourself. Write that API endpoint, design that error state. Less collaboration means less communication overhead for the team and more ownership for each team member. Be the driver.
You may feel like you have to ask someone's opinion before making a decision because "it's not your job". Then you are waiting on an answer. Then you disagree, and have a discussion. Whoever has the biggest ego wins, and now you've wasted four working days.
Learn design
AI still sucks at design. You can make it copy other designs. But to know which designs to copy requires taste — knowing what looks good, what doesn't, and why. Refactoring UI is the best resource out there to teach developers that.
Now that you have taste, use Magic Patterns to brainstorm design ideas using the /inspiration command.
In your coding assistant, install impeccable commands that will point out design issues and help you fix them. I usually start with /critique which tells me what's wrong and which commands to run next like /polish and /normalize.
Be your own manager
Ask yourself, what should I work on? What's the next best thing to do to push the business forward? In an era where we can use agents to build anything, what's left is deciding what to build and how.
Going back to the first point of shipping faster — your team lead can ship faster too. What's stopping them from using Claude instead of delegating to you?
You need to come up with your own ideas, find problems to solve, and prioritize. You're now a PM with agents as your employees. This is a skill that can be learned.
Don't be distracted by the tool
Every week, a new AI tool comes out that tempts you to change your workflow because it "changes the game" and now "we're cooked". Stay curious and keep learning, but always remember that your #1 job is to ship, and AI is just a tool to do that faster and better.
As a side effect, work feels more fulfilling
To sum it all up, you need to have a product mindset (know what to build) and take ownership (build the thing from start to finish).
When you can ship features on your own that are also your own ideas, you can say, "I did that." And that makes you feel more creative and less like a machine ticking off a todo list. You can be proud of your work because it's yours.
Sign in to highlight and annotate this article

Conversation starters
Daily AI Digest
Get the top 5 AI stories delivered to your inbox every morning.
More about
claudeproductcompany
The new Storm Radar app is a treasure trove of data for weather nerds
The horrible winter of 2026 is behind us in New England; now we’ve moved on to the season where there’s a threat of rain basically every day. Given that, the updated Storm Radar app from The Weather Company (owners of the Weather Channel app) caught my eye. There are tons of good weather apps out there, and I’m the kind of nerd that likes to try them all, and Storm Radar feels pretty unique to me. The main interface is, as you’d expect, your local radar. Tapping on any point of the map calls up a detailed forecast for that exact point, with data coming via The Weather Company’s “gridded forecast on demand” (FOD) system. The default view is precipitation, but there are multiple other layers you can add including temperature, cloud cover, nearby lighting strikes, wind and so forth. There are

The end of 'shadow AI' at enterprises? Kilo launches KiloClaw for Organizations to enable secure AI agents at scale
As generative AI matures from a novelty into a workplace staple, a new friction point has emerged: the "shadow AI" or "Bring Your Own AI (BYOAI)" crisis. Much like the unsanctioned use of personal devices in years past, developers and knowledge workers are increasingly deploying autonomous agents on personal infrastructure to manage their professional workflows. "Our journey with Kilo Claw has been to make it easier and easier and more accessible to folks," says Kilo co-founder Scott Breitenother. Today, the company dedicated to providing a portable, multi-model, cloud-based AI coding environment is moving to formalize this "shadow AI" layer: it's launching KiloClaw for Organizations and KiloClaw Chat, a suite of tools designed to provide enterprise-grade governance over personal AI a
Knowledge Map
Connected Articles — Knowledge Graph
This article is connected to other articles through shared AI topics and tags.
More in Products

The new Storm Radar app is a treasure trove of data for weather nerds
The horrible winter of 2026 is behind us in New England; now we’ve moved on to the season where there’s a threat of rain basically every day. Given that, the updated Storm Radar app from The Weather Company (owners of the Weather Channel app) caught my eye. There are tons of good weather apps out there, and I’m the kind of nerd that likes to try them all, and Storm Radar feels pretty unique to me. The main interface is, as you’d expect, your local radar. Tapping on any point of the map calls up a detailed forecast for that exact point, with data coming via The Weather Company’s “gridded forecast on demand” (FOD) system. The default view is precipitation, but there are multiple other layers you can add including temperature, cloud cover, nearby lighting strikes, wind and so forth. There are

The end of 'shadow AI' at enterprises? Kilo launches KiloClaw for Organizations to enable secure AI agents at scale
As generative AI matures from a novelty into a workplace staple, a new friction point has emerged: the "shadow AI" or "Bring Your Own AI (BYOAI)" crisis. Much like the unsanctioned use of personal devices in years past, developers and knowledge workers are increasingly deploying autonomous agents on personal infrastructure to manage their professional workflows. "Our journey with Kilo Claw has been to make it easier and easier and more accessible to folks," says Kilo co-founder Scott Breitenother. Today, the company dedicated to providing a portable, multi-model, cloud-based AI coding environment is moving to formalize this "shadow AI" layer: it's launching KiloClaw for Organizations and KiloClaw Chat, a suite of tools designed to provide enterprise-grade governance over personal AI a

Discussion
Sign in to join the discussion
No comments yet — be the first to share your thoughts!