Weekly Digest
Agents, Enterprise AI, and Image Prompts
Jan 15 – Jan 21, 2024
A diverse week covering AI agents with function calling, enterprise deployment risks, and Midjourney prompt engineering.
Top Tips This Week
All Tips from This Week
How do I write better Midjourney prompts?
Structure Midjourney prompts as: subject + style + lighting + camera angle + quality modifiers. For example: "portrait of a scientist, oil painting style, golden hour lighting, close-up shot, highly detailed, 8k". Order matters—put the most important elements first.
What are negative prompts in image generation?
Negative prompts tell the model what NOT to include. In Stable Diffusion, add things like "blurry, low quality, distorted hands, extra fingers" to improve output quality. This is especially useful for avoiding common AI image artifacts.
How do I build a simple AI agent with function calling?
Define your tools as JSON schemas, then let the model decide when to call them. Start simple: one tool for search, one for calculation. The model returns a function call instead of text, you execute it, feed results back. Loop until done.
What is the ReAct pattern for AI agents?
ReAct combines Reasoning and Acting. The agent thinks out loud (Reason), takes an action (Act), observes the result, then repeats. Format: "Thought: I need to... Action: search(...) Observation: [results] Thought: Now I know..."
How should I evaluate AI outputs in my organization?
Create a rubric before deploying. Define what "good" looks like with specific criteria: accuracy, tone, completeness, safety. Have humans rate a sample of outputs weekly. Track trends over time. Don't just measure speed—measure quality.
What are the key risks of enterprise AI deployment?
Three main risks: data leakage (sensitive info in prompts going to third parties), hallucination liability (AI confidently stating false information), and shadow AI (employees using unapproved tools). Address each with policies, not just technology.