Raycast Launches Glaze for App Creation
Raycast just shipped a new product called Glaze. It allows you to build real applications by describing them in natural language. You provide the idea in plain English. The AI handles the underlying code. This is a major step toward a future where software is built by explaining it, not just by writing it line by line. It's a practical application of prompt-driven development.
For those unfamiliar, Raycast is a popular launcher for macOS. It acts like a supercharged command center for your computer. Developers and knowledge workers use it to create shortcuts and automate tasks. Glaze extends this core idea of productivity. Instead of just controlling existing apps, you can now create new ones on the fly. This moves Raycast from a tool that optimizes workflows to one that creates them from scratch.
The launch announcement showed several examples. A user could type, "Create an app to check the stock price of a company." Glaze would then generate a simple interface with an input field for the stock ticker and a display for the price. This isn't just a code snippet generator. It builds a complete, usable application that lives within the Raycast environment. The term "vibe coding," once a meme, is now a legitimate product category.
This approach lowers the technical barrier to creation significantly. It suggests a world where the primary skill is not writing code, but clearly defining a problem and its desired solution. The focus shifts from syntax and libraries to logic and communication. Raycast is betting that the next wave of creators won't all be traditional software engineers. They will be domain experts who can finally build the tools they've always needed.
What This Means for Your Career
This technology will reshape roles and skill values. For non-engineers, the change is immediate and empowering. Product managers, designers, and entrepreneurs can now build their own minimum viable products. They no longer need to wait for engineering resources to test a hypothesis. This makes the process of Product Discovery much more direct and iterative. An idea from a morning meeting can become a working prototype by the afternoon.
This creates a new class of builder. Someone who understands a business problem deeply and can translate it into a functional tool using AI. Their value isn't in knowing JavaScript. It's in knowing the customer and the workflow. These individuals can now solve their own problems without needing to file a ticket. This is the evolution of tools like Zapier or Bubble, moving from visual programming to pure language. The skill of Low-Code App Building (Bubble/Retool) is merging with the skill of clear communication.
For software engineers, this is not an extinction event. It's a role evolution. The demand for writing routine, boilerplate code will decrease. AI is getting very good at that. Instead, the value will shift to more complex, abstract tasks. An engineer's job will be less about implementation details and more about high-level System Architecture. They will be the ones designing the complex frameworks that AI-generated components plug into. Their role becomes ensuring the foundation is solid, secure, and scalable.
The most effective engineers will become AI orchestrators. They will be experts at guiding AI to produce correct, efficient, and secure code. This requires a different set of abilities. It's about writing excellent specifications, reviewing AI-generated code for subtle flaws, and debugging systems that are part human, part machine. Mastering the art of the perfect instruction, or Prompt Engineering, becomes a core competency for developers, not just writers or artists. The best developers will be the best communicators.
What To Watch
Raycast's Glaze is not an isolated product. It is a signal of a much broader industry shift. We can expect to see similar natural language interfaces appear in all major development platforms. GitHub Copilot, for example, already writes code based on comments. The next logical step is for it to build entire features or applications from a high-level description. The IDE of the future might look more like a chat window than a text editor.
In the short term, the primary use case will be for internal tools and simple, single-purpose apps. Think about the countless small workflow gaps inside every company. The custom scripts and forgotten spreadsheets. Tools like Glaze can fill these gaps quickly and cheaply. This will unlock a huge amount of productivity that was previously blocked by limited engineering bandwidth. We'll see a Cambrian explosion of bespoke, internal software built by the teams that actually use it.
Looking further ahead, this trend will fundamentally alter how we teach and value technical skills. A computer science curriculum might spend less time on language syntax and more time on formal logic, system design, and clear communication. The definition of a "technical" person will expand. It will include anyone who can clearly articulate a process and use AI to manifest it as software. The future of building is not about learning the machine's language. It's about teaching the machine to understand ours.
This also introduces new challenges. How do we manage security for dozens of AI-generated internal apps? What happens when a business process relies on a tool that no single person knows how to code? Companies will need new governance and maintenance strategies for this new class of software. The roles of IT and security will have to adapt just as quickly as the roles of builders. The next frontier is managing the output of this new creative explosion.