DeepMind paper turns 10GB chip waveform dumps into SQL for AI agents, hits 95.33% accuracy
rohanpaul_ai · x · 2026-11-16
- Google DeepMind and Columbia University introduced Back-to-the-Future (BTTF), an agentic framework for AI-driven chip design verification.
- Chip simulations record every signal change, producing waveform dumps often exceeding 10 GB of text that engineers still scan manually; 74.6% of LLM-in-EDA research targets static RTL code generation, leaving post-simulation debugging largely untouched.
- BTTF distills massive unstructured simulation dumps into a normalized SQLite database and couples it with a multi-agent orchestration engine that translates natural-language verification queries into schema-aware SQL, correlating signal anomalies with versioned RTL repositories.
- It achieves 95.33% execution accuracy on a 150-query benchmark; accepted at the NeurIPS'26 Workshop on AI for Chip Design.
More from coding & agent
- DeepMind paper: chip sim waveforms as SQLite lets agents ace 95.33% of verification queries — rohanpaul_ai · 2026-11-16
- 30% of Audited Lovable-Built Apps Expose Sensitive Data, Security Firm Finds — emmanuelvivier · 2026-11-16
- Anthropic Launches Claude for Manufacturing With 17 Skills and 4 Connectors — emmanuelvivier · 2026-11-16
- Building an autonomous game-clip pipeline with Codex without burning usage limits — gondorzabitasi · 2026-11-16
- llama.cpp already supports /v1/realtime for local speech-to-speech chat with LFM3-Omni — ngxson · 2026-11-16
- AI agent swarm of 20 Claude models proves first nontrivial lower bounds for Naslund's crossing problem, formalized in Lean — burny_tech · 2026-11-16