Getting Started with AI-Powered Features in Your App

The AI Integration Landscape

Adding AI to your application no longer requires a machine learning background. Modern AI APIs expose powerful capabilities through simple HTTP calls. The challenge is now deciding what to build and how to integrate it well.

Common Use Cases

• Content generation: Drafting, summarizing, translating
• Classification: Sentiment analysis, moderation, categorization
• Extraction: Parsing structured data from unstructured text
• Embeddings: Semantic search, recommendations, deduplication

Structured Output

For production use, avoid free-form text output. Ask the model to return JSON and validate it:

const result = await client.messages.create({
  model: "claude-opus-4-6",
  max_tokens: 512,
  messages: [{
    role: "user",
    content: `Extract the following from this job posting as JSON:
{ "title": string, "company": string, "salary": string | null }

Job posting: ${jobText}`
  }]
});

const data = JSON.parse(result.content[0].text);

Streaming Responses

For long outputs, stream the response to improve perceived performance:

const stream = await client.messages.stream({
  model: "claude-opus-4-6",
  max_tokens: 2048,
  messages: [{ role: "user", content: prompt }],
});

for await (const chunk of stream) {
  if (chunk.type === "content_block_delta") {
    process.stdout.write(chunk.delta.text);
  }
}

Handling Failures

AI APIs can be slow and occasionally fail. Build resilient integrations with timeouts, retries, and graceful fallbacks:

async function withAIFallback<T>(
  fn: () => Promise<T>,
  fallback: T
): Promise<T> {
  try {
    return await Promise.race([fn(), timeout(10_000)]);
  } catch {
    return fallback;
  }
}

Cost Management

Tokens cost money. Cache responses where possible, use shorter prompts, and choose the smallest model that produces acceptable quality for your use case.