Long-term Memory for AI. The Pinecone vector database makes it easy to build high-performance vector search applications. Developer-friendly, fully managed, and easily scalable without infrastructure hassles.
Deletes one or more vectors by ID, from a single namespace. See the documentation.
Create a new document in a collection of your choice. See the docs here
Looks up and returns vectors by ID, from a single namespace.. See the documentation.
Searches a namespace, using a query vector. It retrieves the ids of the most similar items in a namespace, along with their similarity scores. See the documentation.
The Pinecone API enables you to work with vector databases, which are essential for building and scaling applications with AI features like recommendation systems, image recognition, and natural language processing. On Pipedream, you can create serverless workflows integrating Pinecone with other apps, automate data ingestion, query vector databases in response to events, and orchestrate complex data processing pipelines that leverage Pinecone's similarity search.
import { axios } from "@pipedream/platform"
export default defineComponent({
props: {
pinecone: {
type: "app",
app: "pinecone",
}
},
async run({steps, $}) {
return await axios($, {
url: `https://api.pinecone.io/collections`,
headers: {
"Api-Key": `${this.pinecone.$auth.api_key}`,
},
})
},
})
The MongoDB API provides powerful capabilities to interact with a MongoDB database, allowing you to perform CRUD (Create, Read, Update, Delete) operations, manage databases, and execute sophisticated queries. With Pipedream, you can harness these abilities to automate tasks, sync data across various apps, and react to events in real-time. It’s a combo that’s particularly potent for managing data workflows, syncing application states, or triggering actions based on changes to your data.
import mongodb from 'mongodb'
export default defineComponent({
props: {
mongodb: {
type: "app",
app: "mongodb",
},
collection: {
type: "string"
},
filter: {
type: "object"
}
},
async run({steps, $}) {
const MongoClient = mongodb.MongoClient
const {
database,
hostname,
username,
password,
} = this.mongodb.$auth
const url = `mongodb+srv://${username}:${password}@${hostname}/test?retryWrites=true&w=majority`
const client = await MongoClient.connect(url, {
useNewUrlParser: true,
useUnifiedTopology: true
})
const db = client.db(database)
const results = await db.collection(this.collection).find(this.filter).toArray();
$.export('results', results);
await client.close()
},
})