ai-feature-architect

Create a route handler that calls streamText with your model and messages, then return result.toDataStreamResponse(). On the client, use the useChat hook from ai/react to manage streaming state, input handling, and loading and error states.

Use generateObject with a Zod schema defining the expected shape. The SDK validates the model output against the schema and returns a fully typed object, so you never need unsafe casts like as unknown as Type.

Chunk documents, embed them in batch with embedMany, and store vectors in a PostgreSQL pgvector column. At query time, embed the user query with embed, run a cosine similarity search, and inject the top chunks into the streamText system prompt.

Yes. Define tools with the tool() helper, providing a description, Zod parameters schema, and an execute function. Pass them to streamText and set maxSteps to cap how many tool calls the model can make before producing a final response.

Set maxTokens on every streamText or generateText call, add per-IP and per-user rate limiting, and log token usage in the onFinish callback. For tool calling, always set maxSteps to prevent infinite tool-call loops.

Use routing when request complexity varies: send simple queries to a cheap fast model like gpt-4o-mini and complex or reasoning tasks to stronger models like claude-sonnet or claude-opus. Classify each request first, then select the model accordingly.