SDK & API
A unified API for AI development
Waveloom simplifies AI complexity with a single interface for every model, storage service, and vector database.
with-waveloom.js
With Waveloom
// One import (TypeScript SDK coming soon)import { Waveloom } from 'waveloom';const waveloom = new Waveloom({ apiKey: process.env.WAVELOOM_API_KEY });const result = await waveloom.flows.run({flowId: "my-content-workflow",});console.log(result.output);
before-waveloom.js
Before Waveloom
// Multiple import statementsimport { OpenAI } from 'openai';import Anthropic from '@anthropic-ai/sdk';import { PineconeClient } from '@pinecone-database/pinecone';import { S3Client } from '@aws-sdk/client-s3';// Managing connectionsconst openai = new OpenAI({ apiKey: process.env.OPENAI_API_KEY });const anthropic = new Anthropic({ apiKey: process.env.ANTHROPIC_API_KEY });const pinecone = new PineconeClient();await pinecone.init({environment: process.env.PINECONE_ENV,apiKey: process.env.PINECONE_API_KEY});// Different libraries for each serviceconst openaiResponse = await openai.chat.completions.create({model: "gpt-4",messages: [{ role: "user", content: prompt }]});// Separate auth and api keys, costs for each providerconst anthropicResponse = await anthropic.messages.create({model: "claude-3",max_tokens: 1000,messages: [{ role: "user", content: prompt }]});// Different response formats for each providerconst imageResponse = await openai.images.generate({prompt: "A beautiful sunset over a calm ocean",n: 1,size: "1024x1024"});// Manual error handling and retriestry {const embeddings = await embedding.embed(text);await vectorDB.upsert(embeddings);} catch (error) {// Complex error handling}// Scheduling and monitoringqueue.addJob("my-content-workflow", {input: userInput,userId: "123"});...
Model Flexibility
Switch between providers without changing your code. Use GPT-4 today, Claude tomorrow.
Integrated Storage
All file handling, vector databases, and media processing through one consistent API.
Deploy Instantly
No infrastructure management. Flows you build deploy to production in seconds.
Developer Toolkit
Utilize our comprehensive REST API or SDKs (coming soon) for seamless workflow management and execution directly within your application.
app.js
config.js
1
2
3
4
const result = await waveloom.flows.run);
console.log);
Importing SDK
Running workflow
Success: Workflow completed
Event-Driven Triggers
Trigger AI workflows via API events or external system webhooks, and send real-time notifications upon completion. Integrate Waveloom seamlessly into your existing applications and services.
API Call / Event
Waveloom Workflow
Triggered
{ event: 'image_uploaded', ... }
{ event: 'new_article', ... }