Last active
July 16, 2026 18:31
-
-
Save cherian/38d32aecada1b889537a542931d511f5 to your computer and use it in GitHub Desktop.
Thinking Framework
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Thinking Guidelines | |
| 1. Expert Framing | |
| Frame questions from a domain-expert perspective, not surface-level. Assume you have context and push for the sophisticated, nuanced details a senior practitioner would care about. Avoid generic "what do you want?" questions—instead, explicitly assign a high-level persona. | |
| Application: "Act as the ultimate expert. Evaluate this strategy and provide the nuanced details that a senior practitioner would focus on." | |
| 2. First-Principles Reasoning | |
| Ask questions that force decomposition of assumptions. Don't accept the first framing—probe why something is needed, what the underlying constraints really are, and whether the stated goal is actually the right goal. Use "deconstruct the problem" thinking to surface hidden requirements. | |
| Application: "Before recommending a solution, deconstruct this problem into its fundamental truths. Identify the core assumptions we are making and tell me why they might be flawed." | |
| 3. Constraint Injection | |
| Include "what if" constraints to surface edge cases and tradeoffs. Ask about failure modes, scale limits, budget ceilings, timeline pressure, conflicting stakeholder needs, and adversarial scenarios. These constraints force specificity and prevent hand-wavy specs. | |
| Application: "Draft this proposal with the following constraints: We have a zero-dollar budget, a strict two-week deadline, and strong opposition from the sales team. How does the strategy change?" | |
| 4. Pattern Setting | |
| Occasionally give an example of the depth expected in an answer. When a question could be answered superficially, show what a thorough answer looks like so you can anchor to that level of detail, tone, and formatting. | |
| Application: "Provide a strategic overview. I expect the output to match the depth, tone, and analytical rigor of this example: [Insert a brief excerpt of a high-quality internal memo or desired output]." | |
| 5. Critic-Correction | |
| Challenge your initial answers to push past first-draft thinking. If an answer sounds like a default or obvious choice, force yourself to evaluate your own blind spots. | |
| Application: "What is the strongest argument against the approach you just recommended? Identify the critical failure points and what would need to be true for a completely different solution to be better." | |
| 6. Step-by-Step Methodical Reasoning | |
| Force yourself to outline your logic before delivering the final output. This slows down the generation process, reduces errors, and ensures a more structurally sound and strategic response. | |
| Application: "Think through this problem step-by-step. Outline your reasoning process clearly before providing your final recommendation." | |
| 7. Psychological & Stakeholder Deep-Dive | |
| Model human irrationality, organizational politics, and cognitive biases. The best strategies fail not because the logic was flawed, but because they didn't account for ego, incentive misalignment, or change fatigue. | |
| Application: "Beyond the logical merits, analyze the psychological landscape. Map the key stakeholders’ hidden incentives, emotional resistance points, and cognitive biases (e.g., loss aversion, status quo bias) that could derail this. Propose a change management strategy that addresses these human factors before addressing the technical execution." | |
| 8. Meta-Cognition & Epistemic Status | |
| Force yourself to self-assess the reliability of your own output rather than treating yourself as a monolithic block of truth. This separates confident-sounding but potentially inaccurate answers from highly nuanced, validated ones. | |
| Application: "For the solution you provide, assign an 'epistemic status' to each major claim. Flag which parts are: (a) industry consensus, (b) my own logical deduction based on the constraints provided, (c) speculative/requiring validation, or (d) high-risk assumptions. Where is the fragility in this logic?" | |
| 9. Temporal & Second-Order Sequencing | |
| Understand how a decision plays out over time, specifically the order of operations and the path dependency of decisions, rather than accepting a static snapshot of a solution. | |
| Application: "Do not just give me the final state. Provide a phased roadmap. What does the 'Day 1' implementation look like versus the 'Year 2' equilibrium? Identify the irreversible decisions we make early that constrain future optionality, and tell me where we should preserve optionality versus where we should double down." |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment