AI Writes All the Code, Promotion Passes Him Over: A Chinese Frontend Dev's 2026 Pivot
Think, Act, Review | 2026
Several planned things progressed in 2026: I went through a promotion review, experimented with a side business, adjusted my understanding of business, and redefined my technical direction. Some things I expected results from didn't come through; others exceeded my expectations. By the end of the year, my grasp of business, technology, and commerce had changed considerably compared to the start.
The Promotion Process
I've been at the company for three years with consecutive excellent performance reviews. That resume has some foundation, but I understand that performance alone isn't enough to support a promotion. In March, I took the initiative to report to my supervisor and the regional director, clearly stating my intention to be promoted and laying out the achievements of the past three years one by one.
The feedback wasn't unexpected: I could attend a promotion review, but not get direct approval.
I attended the review in May and failed. Seeing the list, to be honest, my first reaction was a bit defiant. The list had 3 back-ends, 0 front-ends, 0 testers, 3 implementation, and 2 product managers. Not a single front-end. I sat there thinking for a long time, feeling the result didn't add up.
This bothered me for about a month before I slowly understood. A promotion is never a single-dimension comparison; it involves many factors: your relationship with leadership, personal ability, current grade, and how much the company actually needs that role. Technical skill is just one item, possibly not even high on the priority list.
Understanding this wasn't comfortable. My old belief that "good tech gets recognized" was knocked by reality. But from another angle, this review was a practical test that made me realize there are several other dimensions to consider.
Exploring a Side Business
Since May, I've worked on a side business. The earliest approach was simple: a WeChat official account plus a traffic-mini-program. At the time, I thought that if the code was decent and the product was built, monetization would follow.
It expanded step by step: I tried Xiaohongshu, WeChat official account, Douyin, Xianyu, WeChat Video, and Juejin, and spent a small amount on paid promotion. I wanted to see which platform would work, and I also felt unwilling to give up, thinking one more channel might succeed.
Reality gave direct feedback. Every platform enforces strict rules on traffic-grabbing marketing. Xiaohongshu banned me several times, and my WeChat official account got limited. After all the effort, I'd often just start to figure things out when the account disappeared. There was never a smooth path, just stumbling forward.
Looking back at the end of the year, I earned just over 4,000 yuan in total. At first I looked down on that small amount, but after seeing the real state of most people's side hustles, I think it's not a bad result, even a decent ending.
More valuable than the money is the change in perspective. I used to think code was valuable, but once I actually did it, I realized writing code isn't the hard part of the monetization chain. The hard part is promotion, attracting users. This dragged me out of the programmer's mindset, and I started to understand how business models work, and I've developed my own methodology for building an account.
The current state is that side-business income is unstable, basically negligible compared to my main job, and not enough to support full investment. For now I don't plan to give it up; I treat it as an interest and work on it when I have time.
Shift in Business Understanding
When I started working, I loved building infrastructure. I felt business wasn't a high priority; if I built a solid technical foundation, changing jobs would give me more confidence. Looking back, that's a typical mindset that treats technology as the core reliance.
Last year I began to value business ability, understanding that business is an important moat for an individual. But it stayed at the cognitive level; most of my daily time went to technical research, with little time for business.
This year, there's a real shift. I gradually increased my time on business, reading related professional books and doing hands-on work. I notice the change in daily work: last year AI assisted with 70% of my code, this year almost all code writing and modification in projects is handled by AI. Coding no longer takes up much of my energy, and I've started thinking about where my core value as a developer lies.
That makes improving business ability feel urgent.
Current Tech Stack and Next Steps
The company's overall stack uses Nest + Vue3 for full-stack applications, and the backend includes Java service modules. As the team fully pushes AI adoption, the new role requires a single person to independently handle a complete business chain. In this architecture, I plan to fill in Java backend project experience during business iterations and thoroughly understand the full frontend-backend chain.
The core product is integrating Agent plugin capabilities, which implement automatic modeling. I'm deeply involved in this feature's launch, accumulating Agent development experience.
My long-term direction is set: deep development in full-stack and Agent. Next I plan to submit resumes, understand the market conditions, and assess my position.
Summary
This year I tried many things. Some expectations were disappointed, and I also gained some unexpected insights. The promotion didn't produce the result I wanted, and my side business only runs at a light level, but both experiences kept me from purely technical perspectives. Technology is just a tool; coding can be done with AI. The ability to understand business, see through commerce, and bring a product to market needs continuous polishing. I'll keep the habit of acting and reviewing, and gradually accumulate along the full-stack and Agent direction.
Top 2 of 5 from juejin.cn, machine-translated. The original thread is authoritative.
Understanding the business is so hard, how do you generate income?
Personal monetization is even harder; taking a salary is more comfortable.
You're giving up just like that?
Better to use Java. Enterprise development is definitely Java. Using NestJS is unrealistic; it's fine for your own side projects.
Our company has this layer. It's not my own project; I'll need to learn Java later anyway.