In a recent LinkedIn post, Ruben Hassid offers a direct approach to optimizing the use of AI tools like Claude, advocating for the deletion of excessive organizational structures and the adoption of a few key habits. Hassid challenges the common tendency to over-organize AI workflows, arguing that such practices can paradoxically slow down productivity and hinder the effective utilization of AI capabilities.
Challenging the Over-Organization of AI Tools
Hassid begins by listing numerous elements that he believes users should eliminate from their AI setup. This extensive list includes prompt libraries, unused skills, unnecessary connectors, lengthy guides, and deeply nested folder structures. The core of his argument is that these organizational layers often become digital clutter, rarely accessed or utilized, thereby negating their intended purpose and adding friction to the user experience.
He highlights the futility of accumulating resources that are seldom revisited. Hassid states:
“Your prompt library. You never open it.
…The chat history you’ll “go back to read.”
…The plugin pack you installed for one task.”
This sentiment underscores a broader point: the digital hoarding of AI-related assets often serves as a form of procrastination or a false sense of preparedness, rather than facilitating actual use.
The Six Essential Habits for Effective AI Interaction
In contrast to the extensive list of items to delete, Hassid proposes a lean framework of just six core habits and elements to retain for maximizing AI efficiency. These habits are designed to foster a more dynamic and results-oriented interaction with AI models.
1. The “About-Me” File
Hassid emphasizes the power of a single, concise “about-me” file. This file, containing essential information about the user’s role, desired tone, and non-negotiables, acts as a foundational instruction set for the AI. According to Hassid, this single document is more effective than extensive prompt templates in helping the AI adopt a specific persona and avoid generic responses.
2. Focus on Goals, Not Steps
A significant point Hassid makes is the importance of providing the AI with the ultimate goal rather than dictating the precise steps to achieve it. He argues that micromanaging the AI’s process leads to suboptimal output. Instead, he advises users to frame their requests broadly, allowing the AI the autonomy to determine the best path forward.
“Give Claude the goal, not the steps.
… The more you micromanage, the dumber the output.”
3. Encourage AI-Driven Questions
Hassid suggests a simple yet effective prompt addition: “Before you answer, AskUserQuestion.” This encourages the AI to proactively seek necessary context from the user, turning the interaction into a more collaborative information-gathering process. As Hassid notes, this allows the AI to pull the required context from the user, acknowledging that users may not always know how to perfectly prompt.
4. Embrace Iterative Refinement
Challenging the reliance on the first AI-generated response, Hassid advocates for a process of iterative refinement. He recommends prompting the AI to critique its own output and then correct it, stating, “Never trust the first answer.” In his view, the initial output is merely a draft, and the real value comes from the subsequent revisions.
“Claude critiquing Claude beats Claude on the first try. The first draft is a draft, not the answer.”
5. Start Fresh Chats for New Tasks
Hassid strongly recommends initiating a new chat session for each distinct task. He explains that long, ongoing chat histories can negatively impact the AI’s performance, leading to slower responses and generic output due to the accumulation of past interactions. A fresh chat ensures the AI operates with a clean slate.
6. Assign the Hardest Tasks to AI
Finally, Hassid encourages users to leverage AI for their most challenging tasks, rather than delegating only simple, quick jobs. By assigning complex problems, dense reports, or time-consuming tasks to the AI, users can achieve significant time savings, potentially freeing up entire afternoons.
Ruben Hassid concludes his post by summarizing his philosophy: a streamlined setup with one core file and six essential habits, eliminating unnecessary folders and complex organization. He also offers a free “about-me” file via his website for those interested in implementing his approach.
📝 About This Content
This article is based on insights shared by Ruben Hassid on LinkedIn.
📅 Originally posted on August 14, 2026 | View original post on LinkedIn →