Karpathy's Tip: Rambling to LLMs via Voice is a Great Prompting Strategy
nptacek · x · 2026-07-22
Andrej Karpathy shares a practical pattern for working with LLMs: having a nice long ramble session.
- The Method: When you're too lazy to type out complex context, switch to voice and just ramble for about 10 minutes in total stream of consciousness. LLMs are highly capable of reconstructing these incoherent rambles and often echo your tangled thoughts much cleaner than how you started.
Another developer added that this concept is similar to why auto-injecting context from a RAG vector database is so beneficial. Even if the retrieved information isn't a perfect match, providing the model with extra bits of context significantly helps it grasp your underlying intent.
Related event: Karpathy's 10-Minute Voice Rambling Trick for Better LLM Context(16 posts)→
More from coding & agent
- Bolt adds stacked team skills that can all trigger from one prompt — damianplayer · 2026-07-22
- A new RL finding says the training harness may induce generalization — inductionheads · 2026-07-22
- Coding agents are making stagnant enterprise software feel alive again — yunta_tsai · 2026-07-22
- Lakuna Router: Making AI Models 'Audition' for Real-World Workflows — Due_Hovercraft6497 · 2026-07-22
- LangChain says dcode added native browser control with one /goal command — LangChain · 2026-07-22
- Puri.li opens a free web search API with a 185M-page index — skillplayed · 2026-07-22