@dotsokt: Harrison Chase, CEO of LangChain, literally explained why your AI agent usually fails because of the context, not the model. In a 23-minute talk, he breaks down the 3 components that make an agent work: 1. The harness 2. The model 3. The context Most people obsess over the model and ignore the other two. He also explains why OpenAI and Anthropic models can handle the same task differently, and how LangChain changes the implementation depending on the model being used. His key rule: The further your task is from what the model was trained to handle, the more important your own harness becomes. Same model. Better context. Better harness. Better agent. Watch it, save and check the link in bio for more ideas on how to make money with ai. #fyp #fy #business #ai #Tech

7592

302

There are no more comments for this video.

To see more videos from user @dotsokt, please go to the Tikwm homepage.