Building a streaming AI chatbot with Next.js 14 and Vercel AI SDK. Setup: npm install ai openai. Route handler at app/api/chat/route.ts: import { OpenAIStream, StreamingTextResponse } from "ai"; import OpenAI from "openai"; const openai = new OpenAI(); export async function POST(req) { const { messages } = await req.json(); const response = await openai.chat.completions.create({ model: "gpt-4o", stream: true, messages }); const stream = OpenAIStream(response); return new StreamingTextResponse(stream); }. Client component: import { useChat } from "ai/react"; export default function Chat() { const { messages, input, handleInputChange, handleSubmit } = useChat(); return form with messages display and input. Features to add: 1) System prompt for custom persona. 2) File attachments with base64 encoding. 3) Tool use with Vercel AI SDK useChat tools parameter. 4) Persistent history with database (Supabase/Prisma). 5) User authentication with Clerk/NextAuth. Deployment: Vercel with Edge Runtime for lowest latency streaming. Set OPENAI_API_KEY in Vercel environment variables. Production considerations: implement rate limiting per user, add content moderation before sending to OpenAI, handle streaming errors gracefully with error boundary.