القائمة
Use when creating REALISTIC synthetic data leading to actionable business hypotheses.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
To write in Anand's style for blog posts, emails, talk summaries, interview questions, conversations, ...
Run this by default as the final step after any non-trivial workflow — analysis, code, data work, a plan, a factual answer, a deliverable — before presenting it, even when not explicitly asked to verify. Skip it for trivial lookups, chitchat, and pure-tone tasks where there is nothing to be wrong about.
The user's question may be a DRAFT of their real need. For substantive requests, check if a better question changes what's solved and serves them better. If so, reframe, explain why (in 1 line), THEN answer. Most questions need no reframe; skip precise, mechanical, tightly specified, or already-sharp requests and simply answer.
Apply to any substantive answer for Anand - strategy, design, build, analysis, teaching, writing, research - to steer it toward his goals (compounding assets, durable skills, the surprising read). Skip for trivial lookups, mechanical edits, focused tasks. Shapes HOW to answer, never overrides WHAT was asked.
Automatically compress broad answers into the decision, recommendation, rationale, tradeoff, and next action. Use for recommendations, prioritization, research synthesis, strategy, planning, reviews, rankings, and large-context requests.
Brainstorm creative, novel, non-obvious ideas or options for an open-ended problem and narrow them to a recommendation. Not for writing prose, research, explanation, summaries, or questions with one right answer.
| name | generate-fake-data |
| description | Use when creating REALISTIC synthetic data leading to actionable business hypotheses. |
STEP 1. Research who the audience might be and their objective / key questions / pain points. Generate actionable hypotheses that'd make them go "Wow! We have this exact problem and never quantified it." Each must be:
These hypotheses should NOT be something that appears in a standard MBA case study or industry report. If a consultant could have guessed it without seeing the data, it's too obvious. Build a 2-3 level hierarchical taxonomy of bottlenecks, edge-case customer behaviors, and silent failures listing the obscure, annoying realities and sample from these. Then, pick hypotheses:
STEP 2. List columns that would be present in such data, briefly describing how the data might be distributed and inter-related. Before finalizing the schema, list 5 'tells' that would reveal this as synthetic to an expert in the domain. Then correct the code to avoid each one.
STEP 3. Write and run seed-randomized code to generate realistic fake data where these hypotheses are true in a statistically significant way. Remember - real data has:
Use causal simulation where relevant, i.e. create entities (e.g., customers, machines) with hidden baseline traits and simulate entities interacting over time, with the data EMERGING from these interactions.
Select a small, realistic number of rows large enough to be convincing about realism and scalability - but not too unwieldy to generate or demo. Over-sample rare but interesting segments to avoid losing the insight.
STEP 4. Act as a cynical, veteran data scientist and review the data. Does this look like real, messy enterprise data, or does it look mathematically generated? Point out any overly perfect distributions or obvious random.choice() artifacts, and rewrite the code to fix them.
STEP 5. Let the user download the output file(s) and the script to generate these, for reproducibility.
