India's data science talent pool is the second-largest in the world, concentrated in domains where the data problems are uniquely complex: BFSI (credit scoring for populations with thin credit files), healthtech (electronic health records in 22 official languages), edtech (learning outcome prediction across India's enormous variation in school infrastructure), and e-commerce (demand forecasting during festival season spikes of 5–10x normal volume). The data analysis prompts on PromptAndSkills are built for practitioners working in these domains, not just for generic data science tutorials. The most downloaded data analysis prompts on the platform fall into two categories: Python/Pandas prompts for exploratory data analysis (used heavily by analysts at startups and mid-size companies who work with messy real-world Indian data), and SQL prompts for business intelligence queries (used by analysts at companies running on PostgreSQL, MySQL, or Redshift). The most common and most costly mistake in data analysis prompting is not specifying the data format, size, and expected output format. A prompt like "analyse this sales data" gives the AI no information to work with meaningfully. A prompt like "I have a Pandas DataFrame with 50,000 rows, columns: order_id (int), order_date (datetime), gmv (float), city (str), category (str). Write a Python function that calculates week-over-week GMV growth by city and category, handles missing values by forward-filling, and returns a pivot table formatted for export to Excel" produces code you can run on the first try.
20 results