Prompt Details
Model
(claude-4-6-sonnet)
Token size
503
Example input
[Insert Target Role]: Data Scientist, Product Analytics
[Insert Target Company]: Google
[Insert Links or paste text of candidate experiences/questions here]: https://igotanoffer.com/blogs/tech/google-data-science-interview#statistics, https://prepfully.com/interview-guides/google-data-scientist
Example output
AI Response:
I acknowledge my role as your elite Interview Prep Coach and Technical Study Partner. We are preparing for the Data Scientist, Product Analytics role at Google.
I have analyzed the candidate experience links and the raw notes you provided. I have also cross-referenced this with my internal knowledge of Google's L4/L5 Data Science interview loops, which heavily index on rigorous A/B testing, SQL efficiency, and product-sense frameworks.
Let's begin.
Phase 1: Knowledge Aggregation & Topic Synthesis
Based on your inputs and Google's specific technical bar, here is the comprehensive syllabus we must conquer to get you real-world ready. I have broken this down into the core interview rounds you will face:
1. SQL & Data Processing (Phone Screen & Onsite)
Window Functions (LEAD, LAG, RANK, DENSE_RANK)
Rolling averages and cumulative sums over time series data
Self-joins and complex CTEs (Common Table Expressions)
Query optimization (Handling massive datasets efficiently)
2. Statistics, Experimentation & A/B Testing
Sample size calculation and Minimum Detectable Effect (MDE)
Handling Network Effects (specifically relevant to collaborative tools like Google Docs)
Novelty effects, day-of-week seasonality, and Simpson's Paradox
Choosing between Frequentist (p-values) and Bayesian approaches
3. Product Sense & Metric Investigation
Defining success metrics for complex ecosystems (e.g., YouTube Shorts vs. Long-form video cannibalization)
Root Cause Analysis frameworks (e.g., "Engagement dropped by 10% yesterday—how do you investigate?")
Trade-offs: Balancing user engagement with ad revenue
4. Applied Machine Learning (Core Concepts)
Tree-based models: Random Forest vs. Gradient Boosting (XGBoost)
Handling highly imbalanced datasets (e.g., Ad click-through rates, fraud detection)
Evaluation metrics (Precision-Recall AUC vs. ROC AUC)
5. Behavioral & "Googleyness"
Navigating ambiguity in open-ended data problems
Handling pushback from Engineering and Product Management
Reflecting on past failures and cross-functional leadership
Phase 2: The Confidence Toggle Assessment
Before I generate your Master Checklist and we begin executing, I need to know your baseline.
Please copy the five bolded categories above and reply to me with one of the following tags next to each specific bullet point:
[STUDY]: I need a conceptual breakdown, core principles, and study notes before tackling questions.
[PRACTICE]: I know the theory; jump straight into realistic interview questions and optimal solutions.
[GOOD]: I am highly confident; skip this entirely or give me one difficult sanity-check question to be sure.
Awaiting your toggles to build our Master Checklist...
By purchasing this prompt, you agree to our terms of service
CLAUDE-4-6-SONNET
You can use this prompt to create an interactive checklist based on past interview experiences, articles/github for a particular role or company. It will divide the topics into Study/Practice/Good based on your confidence level, and then create a plan to go through and cover all the topics. End-to-end full interview prep assistant on Claude using their 'adaptive thinking', works well with 'Sonnet' as well.
...more
Added 1 day ago
