Cite
Notes
Only stored in your browser.
Agentic software-engineering tasks evaluated inside a real IDE development workflow, not isolated patch generation. By AfterQuery.
Full-stack web-application generation from natural-language prompts across six domain tasks (finance, healthcare, legal, and more). By AfterQuery.
LLMs write and backtest quantitative trading strategies (simple trading, pairs trading, dynamic hedging), scored on backtest error and executable passes. By AfterQuery.
Real-world financial-analysis questions sourced from primary documents (10-K filings), graded by exact match. By AfterQuery.