Atomic AI advice extracted from the best creators. Quick tips you can use right now.
This Week
Feb 5 – Feb 11 #AI Agents
This week focused on building reliable AI systems. Key insights on writing persistent system prompts and handling hallucinations in production environments.
Put the most important instructions at the start AND end of the system prompt (primacy and recency effects). Use clear formatting like numbered lists. Repeat critical constraints. Test with adversarial inputs to see if the model breaks character.
Use GPT-4 for prototyping and diverse tasks. Fine-tune smaller models when you have consistent, narrow use cases with thousands of examples. Fine-tuned models are cheaper to run but expensive to train and maintain. Start with GPT-4, only fine-tune if cost becomes prohibitive.
Layer defenses: ask the model to cite sources, implement fact-checking against known databases, flag low-confidence responses for human review, and set clear expectations with users that AI can make mistakes. Never fully automate high-stakes decisions.
Few-shot prompting means giving 2-5 examples before your actual request. Format: "Example 1: [input] -> [output]. Example 2: [input] -> [output]. Now do this: [your input]". The model learns the pattern from examples.
Use Cursor's @codebase mention to give the AI context of your entire project. Then ask for specific refactoring: "Rename all instances of UserService to AuthService and update all imports." Cursor will propose changes across multiple files.
Creator of Datasette, co-creator of Django. Writing about LLMs, prompt engineering, and building with AI.
Professor at Wharton studying AI and innovation. Author of "Co-Intelligence".
AI researcher at Notion. Building tools for thought and exploring human-AI interaction.
AI engineer and educator. Founder of Latent Space, exploring the AI engineering landscape.