Cast AI SDK Patterns

Overview

Production-ready patterns for Cast AI SDK usage in TypeScript and Python.

Prerequisites

• Completed castai-install-auth setup
• Familiarity with async/await patterns
• Understanding of error handling best practices

Instructions

Step 1: Implement Singleton Pattern (Recommended)

// src/castai/client.ts
import { CastAIClient } from '@castai/sdk';

let instance: CastAIClient | null = null;

export function getCast AIClient(): CastAIClient {
  if (!instance) {
    instance = new CastAIClient({
      apiKey: process.env.CASTAI_API_KEY!,
      // Additional options
    });
  }
  return instance;
}

Step 2: Add Error Handling Wrapper

import { Cast AIError } from '@castai/sdk';

async function safeCast AICall<T>(
  operation: () => Promise<T>
): Promise<{ data: T | null; error: Error | null }> {
  try {
    const data = await operation();
    return { data, error: null };
  } catch (err) {
    if (err instanceof Cast AIError) {
      console.error({
        code: err.code,
        message: err.message,
      });
    }
    return { data: null, error: err as Error };
  }
}

Step 3: Implement Retry Logic

async function withRetry<T>(
  operation: () => Promise<T>,
  maxRetries = 3,
  backoffMs = 1000
): Promise<T> {
  for (let attempt = 1; attempt <= maxRetries; attempt++) {
    try {
      return await operation();
    } catch (err) {
      if (attempt === maxRetries) throw err;
      const delay = backoffMs * Math.pow(2, attempt - 1);
      await new Promise(r => setTimeout(r, delay));
    }
  }
  throw new Error('Unreachable');
}

Output

• Type-safe client singleton
• Robust error handling with structured logging
• Automatic retry with exponential backoff
• Runtime validation for API responses

Error Handling

Examples

Factory Pattern (Multi-tenant)

const clients = new Map<string, CastAIClient>();

export function getClientForTenant(tenantId: string): CastAIClient {
  if (!clients.has(tenantId)) {
    const apiKey = getTenantApiKey(tenantId);
    clients.set(tenantId, new CastAIClient({ apiKey }));
  }
  return clients.get(tenantId)!;
}

Python Context Manager

from contextlib import asynccontextmanager
from castai import CastAIClient

@asynccontextmanager
async def get_castai_client():
    client = CastAIClient()
    try:
        yield client
    finally:
        await client.close()

Zod Validation

import { z } from 'zod';

const castaiResponseSchema = z.object({
  id: z.string(),
  status: z.enum(['active', 'inactive']),
  createdAt: z.string().datetime(),
});

Resources

• Cast AI SDK Reference
• Cast AI API Types
• Zod Documentation

Next Steps

Apply patterns in castai-core-workflow-a for real-world usage.