We should be more tired than the model
vickiboykis.com
May 29, 2026
2 min read
🔥🔥🔥🔥🔥
57/100
Summary
Agentic code generation can lead to feelings of losing control over the coding process. Writing code with AI assistance may result in outward signs of completion without engaging the same internal cognitive processes as manual coding.
Key Takeaways
- Agentic code generation can lead to a lack of internal cognitive processes typically involved in manual coding, resulting in feelings of brain fog after sessions.
- The author suggests that deliberate use of AI tools, such as rewriting code and engaging in discussions, can enhance skill retention and understanding in coding.
- Adding friction to the coding process, like spending time on problems before using AI, ultimately improves long-term proficiency with coding tools.
- The experience of using AI for coding can resemble a slot machine, where quick solutions may detract from deeper learning and retention of programming skills.
Community Sentiment
Positives
- LLMs are enabling product managers to focus on higher-level design and requirements, enhancing their skills while delegating coding tasks to agents.
- Using coding agents for extensive refactoring allows developers to manage code quality better without getting bogged down in the details, which can improve overall productivity.
- Active code review while the AI generates small code batches can lead to better understanding and quality assurance, making the development process more efficient.
Concerns
- The introduction of AI tools in software development could lead to increased pressure on developers to produce more, potentially jeopardizing job security and work-life balance.
- Many companies are likely to retain productivity gains for themselves rather than sharing them with employees, which could exacerbate job insecurity in the tech industry.
- There is a concern that LLMs may not provide the necessary abstractions for effective programming, leaving developers struggling with understanding complex systems.
