Supercharge Your Web App with an AI Writing Companion with Q&A
summary
Builds a Tiptap AI extension that delegates prompts to a server-side proxy so OpenAI keys stay off the frontend, uses GPT-3.5-turbo via the OpenAI API, and caches responses for speed and cost control. Features include summarize, translate, change tone, and continue writing with a custom node view (React renderer optional). Demonstration shows translation losing bold styling unless prompts ask to preserve formatting, and explains token limits (4k/8k) and limiting response tokens to avoid overflow.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 1:08AI extension for Rich Text Editors
- 3:20Demo: TipTap AI extension
- 7:15AI in content creation
- 8:44Code walkthrough
- 14:25AI chat completion function
- 18:27Client-server communication
- 19:49Token limits in AI responses
- 23:24Q&A
related talks
32mConversations Without Prompts: AI in the Web
46mPractical End-to-End AI Development using Prompty and AI Studio
01h 08mIn Prompts We Trust
49mBuilding AI Enabled Web Apps
45mBringing the Power of AI to Your Application
42mAdd Useful AI to Your Web App (Not Just Chatbots)
21mUnlocking Client-Side AI in Your Web App
19mGenerative AI for Front-end Developers