The author uses AI tools like GitHub Copilot, Cursor, and Aider, and has developed his own back-end app for constant access to LLMs. He stresses that the value derived from LLMs is not directly related to the model itself, but rather to the way developers work with them. Despite debates on the limitations of LLMs, he focuses on exploring the opportunities they offer and adapting to the changing landscape of software development.
Key takeaways:
- The author uses large language models (LLMs) to generate up to 90% of the code for his projects, changing the way he creates software.
- Despite debates about the intelligence and reasoning capabilities of LLMs, the author focuses on exploiting the opportunities they offer and exploring their potential.
- The author emphasizes the importance of availability, customization, using the right tools, and creativity in effectively using LLMs.
- Despite the potential of AI to replace jobs, the author chooses to focus on areas he can control, such as how he uses current opportunities and explores new ones.