The concept of DIY or personal software is gaining traction, as it becomes economically viable to design software for individual needs, reflecting a shift in software development.

In 2024, we have personal software that allows anyone to build applications quickly, similar to how the Mac democratized personal computing in 1984.

Vibe coding products like Cursor and Bolt are emerging, allowing users to create applications from text prompts, although they have yet to reach mainstream adoption.

AI tools are now good enough to create human-simulating applications, and platforms like SMS and email are becoming more open, presenting new opportunities for startups.

The emergence of partial autonomy apps, like Cursor, illustrates how LLMs can enhance traditional software interfaces by managing context and orchestrating multiple calls. Software is fundamentally changing again, as we enter the era of 'Software 3.0,' where natural language becomes the new programming interface.

In the future, we may see a shift towards on-demand custom software, where users interact with AI to generate specific code solutions tailored to their needs, resulting in ephemeral programs that solve immediate problems. There will also be a demand for smart individuals who can effectively manage AI coding tools, and if we label them as software engineers, that's acceptable, but the nature of the job will be significantly different.

Custom-designed software tailored for individual users may emerge, but there are challenges in achieving this level of personalization effectively.

The goal is to create tools that allow composers to offload interesting aspects of their work, like expressive timing and musical texture.

Mobile applications are thriving in the AI space, particularly those that allow users to edit and splice video on the go, as they cater to the convenience of mobile usage.