Columbia and Google unveil BTTF agentic framework for RTL debugging

Researchers at Columbia University and Google DeepMind published a paper on BTTF, an end-to-end agentic framework that automates post-simulation RTL waveform analysis and debugging. It distills simulation dumps into a normalized SQLite database and uses a multi-agent engine to translate natural-language queries into SQL, reaching 95.33% execution accuracy on a 150-query benchmark.

• BTTF converts simulation dumps into a normalized SQLite database
• A multi-agent engine translates natural-language queries into SQL
• 95.33% execution accuracy on a 150-query benchmark
• 74.6% of AI-in-EDA studies target code generation, not debugging