Develop a sophisticated AI writing assistant that drafts blog posts based on your unique style and then iteratively refines them. This workflow uses a structured grading system, modeled after an AP English teacher, to improve the quality of the AI-generated content.
with Claire Vo
Archive your existing body of work (e.g., all your blog posts) into a vector embeddings database like LanceDB. This creates a searchable resource that the AI can use to learn your unique writing style, tone, and structure.
Create a script that takes a topic or transcript as input. The script should first query your vector database to dynamically analyze your style from relevant past posts. Then, feed this analysis into a detailed prompt to generate a first draft that mimics your voice.
You are an expert blog writer specializing in technology and business content. Based on the style analysis of existing posts, write in a style that: {style_analysis_characteristics}. Write approximately 500 words total. NO section headers or H2/H3 tags - write as continuous flowing prose. LIMIT each PARAGRAPH to at most 2 LONG sentences. Create a blog post with a compelling hook.After generating the draft, use a second prompt that instructs the AI to act as a critical AP English teacher. This prompt should ask the AI to grade the draft based on a detailed rubric.
Instruct the AI grader to provide a letter grade, a numerical score, and specific feedback on elements like the hook, argument clarity, evidence and examples, paragraph structure, conclusion strength, and overall engagement.
You're an experienced English teacher. Here's the letter grade numerical score, and then here are the evaluations, the hook, which, you know, argument, clarity, evidence, and examples, paragraph structure, conclusion, strength, overall engagement.Run the draft through the grading process three times. Use the AI's feedback from each iteration to guide the refinement of the post. This process helps enforce brevity and improve logical flow, even if the AI's grade fluctuates.
Review the final, AI-refined draft. As the author, you provide the essential human element, correcting any stylistic nuances the AI missed and ensuring the final piece truly reflects your voice.
Follow Sahil Lavingia's method for delegating small but impactful maintenance tasks, like fixing an internal Slack report, to the AI agent Devin. This workflow shows how to use simple, clear prompts to fix code and update system prompts without distracting your engineering team.
Learn Gumroad CEO Sahil Lavingia's process for rapidly prototyping a UI with v0, then using the AI agent Devin to autonomously code, implement, and submit the feature for review, turning a multi-day task into a one-hour workflow.
Improve your writing in real-time by using ChatGPT as a first reader. Get instant feedback on structure, clarity, and flow as you write, allowing you to polish your work section by section.
We've made ChatPRD affordable so everyone from engineers to founders to Chief Product Officers can benefit from an AI PM.