7 Ways How We Use AI Is Changing
Sunday, 20 September 2026 · 1 min read · Listen to the episode ↗
This episode explores the transformative ways AI is being utilized, focusing on the shift from prompt engineers to loop engineers, which reflects a growing preference for simplified, integrated user experiences. The discussion highlights the importance of ongoing conversations through mono threads and the role of voice integration in enhancing task management. Experts also address the rising demand for multi-modality in AI and the challenges of creating consistent mental models for shared agents in team environments.
The episode discusses the evolving use of AI, highlighting a shift from prompt engineers to loop engineers, which reflects changing interaction patterns. Users are increasingly favoring simplified and integrated experiences, preferring a unified interface for AI interactions rather than managing multiple agents. This trend towards simplicity is expected to continue, although some experts express concerns about losing specific control over interactions.
The introduction of mono threads allows for ongoing conversations that maintain context, while Codex's context compaction enhances information management over time. Nick Baumann from Codex emphasizes that this new approach is exciting and opens up new product directions for coding agents. Users are now maintaining fewer, longer threads for recurring tasks, which increases the value of these interactions.
Voice integration is becoming essential for agent products, with Danny Graziosi noting that managing tasks through voice feels almost magical. However, Garra Bison points out that voice capabilities must be paired with real work functionalities to be effective. The nature of AI instructions is evolving from simple prompts to more complex goal setting, with loop engineering emerging as a critical mindset for continuous improvement in task execution.
There is a rising demand for multi-modality in AI, driven by cost and efficiency needs, as cheaper models are now capable of performing tasks that were once considered advanced. Shared agents are becoming vital for team-based work, with Mio being recognized as the first AI employee shared by an entire team. However, challenges persist in establishing a consistent mental model for multiplayer AI systems, which is anticipated to be a significant design problem in the near future.
As the industry progresses, there is anticipation for a surge in innovative product experiences in the multiplayer AI space. Teams are advised to wait for the landscape to stabilize before fully engaging with these emerging technologies.
This summary was generated from the episode transcript and can contain mistakes.