TsundereSQL

An SQL course that likes you back.

TsundereSQL is an anime visual novel built around a complete SQL course. Write real DuckDB queries, pass hidden tests, and earn story scenes with three instructors who slowly warm up as your skill grows.

Every SQL exercise runs against DuckDB. Start with your first SELECT and climb through 224 problems across 14 modules toward production-shaped analytics. Real rows come back, hidden tests decide whether the answer holds, and the story advances when your query holds.

The course contains 224 authored problems across 14 modules. 220 are SQL exercises that run and grade against seeded DuckDB datasets. Four are open-ended analyst-judgment cases with a local fallback, so the curriculum does not depend on live AI.

SQL Foundations and Aggregation & GROUP BY, from your first SELECT through trustworthy totals and rates.

Joins & Set Operations, including left, right, inner, and anti joins on real tables.

Subqueries & CTEs for turning a hard question into readable steps.

Window Functions, ranking, QUALIFY, frames, gaps and islands, and sessionization.

Product & Growth Analytics, including retention, cohorts, funnels, and experiment-shaped questions.

Advanced SQL, recursion, JSON, reshaping, data at scale, and data modeling.

Working With Production Data, where dates, NULLs, casts, messy JSON, and inconsistent records matter.

Thinking Like a Senior Analyst, the Growth Desk, and the Interview Gauntlet for messy real-world reasoning.

If you already write SQL, start at the module that hurts. If you are starting from zero, start at the top and let the difficulty climb with you.

Three instructors run the course. Tsumugi is demanding, Aimi is patient, and Nora is precise to a fault.

Their authored routes and scenes are earned through the work. Romance is optional; the curriculum is not.

What it is: a SQL course wrapped in a character-driven story for people who learn better when practice has stakes. Each instructor has a fixed, developer-directed route with its own scenes and ending.

What it isn't: it is not a branching-choice simulator where every decision reroutes the plot. The routes are authored and fixed. It is not a reference manual or a multiple-choice quiz: you write the queries, and the game checks them.

The workspace includes a schema browser, query editor, result grid, and a grader that compares results against hidden test data. Passing means your query produces the correct result, not merely that it matches a reference string.

Tsumugi: SQL Foundations, and the placement check that lets you skip what you already know. She will tell you the query is fine, and then tell you what is wrong with it.

Nora: joins, window functions, and the Growth Desk where the question is money. With Nora, "it works" is where the conversation starts.

Aimi: she keeps the record: every solve, every miss, and the review queue that hands your old answers back on fresh data. She is always happy to see you again.

The game includes optional romance storylines and optional suggestive anime artwork (including swimwear or other revealing, non-nude outfits). Before anything else renders on first launch, a content screen requires an affirmative adult confirmation to continue with suggestive artwork or offers a one-click route that disables it. Suggestive artwork is otherwise on by default after confirmation and can be disabled later in Settings. Romance is enabled by default and can be disabled separately. There is no nudity and no explicit sexual content.