Non-engineering Coding is Increasing – Frustrations are Rising for Engineers

More and more companies are experimenting with AI in ways that are harmful to engineers. In a recent Blind survey, professionals confirmed that their employers are handing coding assignments to non-engineers. 

Can anyone be an engineer with AI?

32% of the respondents of this survey said, “non-engineers are overestimating what AI can actually do.” Vibecoding is taking over at companies where the desire for quick code and apps is outpacing the work capabilities of engineers.

One user on Blind shared, “I find it a bit concerning how little some of these people know about the actual code that is going into building what they want to build.” This aligns with the 27% of respondents who said non-engineers should only build prototypes, not production systems.

The minority of 24% thinks AI is making technical work more accessible. With the right learning tools, this can be a perk. Without, however, coding faces risks of bugs and security vulnerabilities.

Do engineers need to know how to code anymore?

A poll on Blind showed that 46% of people think that, as a result of AI, companies need more engineers, not less. Concurrently, 37% of the respondents of this survey said AI speeds up prototyping, but strong engineers become more valuable.

The survey asked participants their predictions of what AI coding tools will do to engineering teams over the next three years. There seems to be conflicting opinions on what the future holds:

• 22% said smaller engineering teams can ship more with AI
• 21% said engineers become reviewers and infrastructure guardrails
• 20% said companies will regret shipping too much low-quality AI code

Some tech professionals think AI killed software engineering. 

Of the professionals who completed this survey, 71% are either engineers or developers.

Methodology

Blind surveyed 1,512 professionals between May 14 to May 26, 2026 on their views of non-engineers using AI to code in their workplace. Response breakdowns by question are as follows:

1. What is your perspective on non-technical employees using AI to code?
   1. AI is making technical work more accessible (24%)
   2. Non-engineers should only build prototypes, not production systems (27%)
   3. Most AI-generated code creates hidden technical debt (18%)
   4. Non-engineers are overestimating what AI can actually do (32%)
2. What will AI coding tools do to engineering teams over the next 3 years?
   1. Engineers become reviewers and infrastructure guardrails (21%)
   2. Smaller engineering teams can ship more with AI (22%)
   3. AI speeds up prototyping, but strong engineers become more valuable (37%)
   4. Companies will regret shipping too much low-quality AI code (20%)
3. Which best describes your role?
   1. Engineer / Developer (71%)
   2. Product Manager (7%)
   3. Designer / Research (3%)
   4. Data / Analytics (6%)
   5. Operations / Business (7%)
   6. Other non-engineering role (6%)

Join your coworkers on Blind.