A Reddit post titled “After Working on LLM Apps, I’m Wondering: Are they really providing value” reflects the author’s skepticism about the advantages of LLM-based applications compared to traditional automation tools. They note that LLM apps primarily process text inputs to determine user intent and call appropriate functions, which doesn’t seem significantly different from previous methods. They express a lack of observed improvements in efficiency or capability.
In the comments, users share diverse perspectives. Some see potential in integrating LLMs into existing software, while others highlight the narrow applicability of these tools. A few commenters emphasize the importance of user experience (UI/UX) and the necessity of a human-in-the-loop approach for effective outcomes. Others discuss the challenges of developing LLM applications, including dealing with infinite user inputs and the need for clear prompts. Some users note that LLMs can enhance productivity in specific contexts, particularly in education and data processing, while others caution that expectations may be too high.
Key takeaways from the discussion include the critical role of UI/UX, the shift in user roles towards solution engineering, and the potential of LLMs as powerful tools when applied thoughtfully to appropriate problems. The conversation concludes with a recognition that while LLMs can offer significant benefits, their value is context-dependent and requires careful integration into workflows.