All video posts

Software development · AI · Career

Why Most Developers Will Get Left Behind (And How to Stay Ahead)

For the first time in years, being a software engineer does not feel safe. Layoffs, juniors competing with seniors, more outsourcing, and now AI writing code faster than any of us. If your value stops at "I code", you are in trouble. Here is how the game is changing and the four moves I am making to stay ahead.

Deric YeeDeric Yee 16 November 2026 6 min read 8 min video

Prefer to read? The full article is below. Watch on YouTube.

The short version

  1. 1AI is great at generating code and bad at maintaining it. Debugging, edge cases and production systems still need real developers.
  2. 2Think of yourself as a business: your code is your product and your employer is your biggest client.
  3. 3Stack development with other skills, especially distribution and systems thinking.
  4. 4Use AI as leverage, like an intern. Just do not let the intern run the company.
  5. 5Build a personal brand. People cannot work with you if they do not know you exist.

We are not in the same market any more

Hundreds of thousands of tech workers were laid off across 2023 and 2024. Juniors are competing with seniors. Mid-level developers are grinding LeetCode and HackerRank again. Outsourcing has ramped up from Latin America to Eastern Europe to Southeast Asia.

But that has been happening for years. The scary part is AI. From first principles, a software engineer's original value was "can you write code?" That value has been commoditised. Claude, ChatGPT, GitHub Copilot and Cursor write code faster than any of us. If that is where your value stops, you are basically in trouble.

The nuance is that AI is great at generating code and bad at maintaining it. You still need real developers to debug, optimise and ship production-ready systems, and when things break, someone has to fix them. Software is never perfect from the first version. There are pivots, tweaks and feature requests, and every one of them needs someone to maintain and build on what exists.

Messy is still where humans win

I have lived this. While building one of our own products and delivering larger software projects for clients, I worked with a lean, fast development team. What slowed us down most was not writing code. It was maintenance: weird edge cases nobody planned for, unexpected user behaviour, browser issues. Things neither we nor Copilot could predict before real testing.

So when people say AI is about to replace developers, I find it unrealistic. I have built real products and I know how messy it gets, and messy is still where humans win. I go deeper into this question in will AI replace developers.

I am also optimistic, because cycles are normal. Every market cools after a hype phase. Crypto saw it, property saw it, and tech is going through its own version. What comes after is usually opportunity: less noise, fewer tourists, and more people who actually want to build something productive.

“AI is not your competition. It's your intern. Just don't let the intern run the company.”

Deric Yee, in the video

Move 1: Think like a business

If you are a software developer, you are not just an employee. You are a business. Your code is your product and your employer is your biggest client.

So I stopped asking "how do I do my job?" and started asking "how do I increase the return on my work?" I track deliverables in tools like Notion and Coda as if they were a product roadmap, and I run weekly retrospectives on myself. One question I ask: if I were billing myself $100 an hour, would I be happy with what I produced today? That shift alone made me at least twice as effective.

This framing also changes how you talk to employers. Instead of listing tasks you can do, you show the outcomes you produce. That is exactly what hiring managers look for in junior developers.

Move 2: Stack high-leverage skills

Everyone wants to be a React developer. Congratulations, you and a million others. What most people do not do is combine development with business, marketing or systems thinking. In business there are really two things that matter: product and distribution.

01

Product

Coding covers most of it. I added some design so I could build good user experiences, not just working features.

02

Distribution

Right after learning to code, I started learning distribution. It is not just SEO or paid ads; it is how things spread. If you can build and distribute, you will do very well in this industry.

03

Systems thinking

I invested in understanding systems design, not to pass an interview, but because real products need real architecture. It is the difference between a code monkey and a systems thinker.

Knowing your limits is part of systems thinking

I will be honest: I am not the best at systems design. For new projects I bring in senior software architects to set up the infrastructure early, and then my developers build on it. For later projects with a similar scope, we reuse a similar architecture. Most web products are relatively similar under the hood; you are mostly doing CRUD, create, read, update and delete.

The point is not to know everything. It is to understand enough to make good decisions, recognise what good architecture looks like, and know when to bring in someone better. That is what separates someone who just writes features from someone who can be trusted with a whole product.

Move 3: Use AI for leverage, not just code generation

Everyone talks about using ChatGPT or Claude to generate code. That is level one, the bottom of the barrel. Here is how I use it instead. I turn messy Notion pages into technical documentation. I simulate code reviews before I push to GitHub. I even built an internal bot that answers questions about our own codebase, using our repository and docs with tools like LangChain.

That is the real leverage. Treat AI as your intern. You have to know how to direct it, check its work and give it the right context. But you stay accountable for what ships. If you want to learn this properly, start with our guide on how to learn coding with AI.

Move 4: Build your personal brand

The most underrated asset is your reputation. I built a personal website not to brag, but because in a world full of noise, clarity is currency. If someone wants to know me, they can spend five minutes reading about who I am before scheduling a call or proposing a collaboration.

The quality of people reaching out improved dramatically. They knew what I was working on, what I stood for and where I was heading, so conversations became far more productive. Clients, partners and collaborators now find me through the content I put out on LinkedIn and YouTube.

In a market where CVs look the same and AI can write anyone's cover letter, your story and your voice are the moat. Start writing and sharing. People cannot work with you if they do not know you exist or what you stand for.

Adapt, do not wait

You can sit around hoping the market improves, or you can adapt. I chose to adapt: to treat my career like a startup, stack skills, use AI as leverage instead of a crutch, and tell my story even when there is fear of being judged.

I truly believe the developers who take ownership like this will not just survive the next phase of tech. They will lead it.

If you are early in your journey and want to build those skills together, fundamentals, AI tooling and real projects, our AI-Native Software Development Programme is built around exactly that. You can test whether it suits you first with the free trial.

Questions people ask

  • Will AI make software developers obsolete?

    Not the ones who do more than write code. AI is very good at generating code quickly, but real products need debugging, maintenance, handling edge cases and unexpected user behaviour, and sound architecture. Those parts are messy and still need human judgement. Developers whose only value is typing code are most exposed; those who solve problems and own outcomes are not.

  • What skills should developers learn besides coding?

    Combine development with skills that multiply its value. Distribution, meaning how products spread through marketing, content and sales, is a big one. Systems thinking and architecture help you build products that scale. Some design helps you create good user experiences. And communication, including a personal brand, helps the right people find and trust you.

  • How should developers use AI beyond generating code?

    Use it as leverage across your workflow: turning rough notes into technical documentation, simulating code reviews before you push, explaining unfamiliar code, and building internal assistants that answer questions about your own codebase. Treat AI like a capable intern: give it context, review its output carefully, and stay accountable for anything that goes into production.

  • Do developers really need a personal brand?

    It helps a lot. When CVs look similar and AI can write anyone's cover letter, a visible track record of what you build and think sets you apart. A simple personal site plus regular posts on LinkedIn or elsewhere means clients, employers and collaborators arrive already understanding your work, which makes conversations far more productive.

Your next step

Our 12-week AI-Native Software Development Programme is a live, mentor-reviewed cohort where you build and ship real products with AI, and learn to explain every line. Start with the free trial to see how we teach, then decide.

Watch next

All videos

Learn the fundamentals that pay off in any direction.
Coding plus AI, taught properly.

Whatever you end up doing, the people who understand how software and AI actually work get to direct them. Try the first lessons free, no card needed, or see the full 12-week AI-Native Software Development Programme.