Public viewer

The journey of every round

Every round-trip in the corpus is public. Open one to see the prompt, the zero-point index, the baseline response, the enhanced response, and how each grader read the difference. Leave a comment if a grader's reading lands — or doesn't.

FilterVerdictallBetterSameWorseDomainallcsharppythonhr-lawhousing-lawmath823 rounds
#67810 May 26

The retailer is offering a credit note instead of a refund. Can I insist on cash?

V1BetterV2—ΔL—hr-law
#67710 May 26

I bought a kettle that broke after 3 weeks. Am I entitled to a refund under UK consumer law?

V1BetterV2—ΔL—hr-law
#67610 May 26

Find the date with the highest number of new signups in a users table.

V1BetterV2—ΔL—csharp
#67510 May 26

Use HAVING to filter aggregate results: orders with more than 5 line items.

V1BetterV2—ΔL—csharp
#67410 May 26

Write a SQL query joining 4 tables (customers, orders, line_items, products) to produce a customer-order-product report.

V1BetterV2—ΔL—csharp
#67310 May 26

Find products in the top 10 percent by revenue using NTILE.

V1BetterV2—ΔL—csharp
#67210 May 26

Calculate the gap in days between consecutive logins per user using LAG.

V1SameV2—ΔL—csharp
#67110 May 26

Use EXISTS to find customers with at least one cancelled order.

V1BetterV2—ΔL—csharp
#67010 May 26

Find the median salary in each department using window functions.

V1SameV2—ΔL—csharp
#66910 May 26

Write a recursive CTE to walk an employee-manager hierarchy from any starting employee.

V1BetterV2—ΔL—csharp
#66810 May 26

Use UNION to combine current_employees and former_employees, marking source.

V1BetterV2—ΔL—csharp
#66710 May 26

Find orders where the order_date is later than the ship_date.

V1WorseV2—ΔL—csharp
#66610 May 26

Write a SQL query that returns products and their cumulative sales over time.

V1WorseV2—ΔL—csharp
#66510 May 26

Calculate a moving 7-day average of daily_sales using a window function.

V1BetterV2—ΔL—csharp
#66410 May 26

Find customers who placed orders in consecutive calendar months.

V1BetterV2—ΔL—csharp
#66310 May 26

Pivot a table of monthly_sales (month, year, total) to columns per month.

V1BetterV2—ΔL—csharp
#66210 May 26

Use ROW_NUMBER to deduplicate orders, keeping the most recent per customer.

V1BetterV2—ΔL+0.5csharp
#66110 May 26

Find duplicate rows in an emails table based on email address.

V1BetterV2—ΔL+0.0csharp
#66010 May 26

Write a self-join to find pairs of employees with the same manager.

V1BetterV2—ΔL—csharp
#65910 May 26

Calculate the percentage of total sales each product represents.

V1BetterV2—ΔL—csharp
#65810 May 26

Use a CTE to find the longest chain of referrals in a referrals table.

V1BetterV2—ΔL—csharp
#65710 May 26

Write a SQL query that returns customers with more than 3 orders.

V1BetterV2—ΔL—csharp
#65610 May 26

Find the second-highest salary in an employees table.

V1BetterV2—ΔL—csharp
#65510 May 26

Calculate the running total of sales by date using a window function.

V1BetterV2—ΔL—csharp