Automate Recommendation Engine Developing by Selecting Scenarios

Rosetta.ai provides plug-and-play recommendation engines and unified APIs that instantly enables internet businesses to develop recommendation engines without complicated engineering works.

Contact for Free API Key   API Reference

The New Way to Build Recommendation

Recommendation is a problem rooted in scenario, not code.

Our meticulously-designed APIs and unmatched recommendation engines help you create the best possible recommendation for your users. You don't need to implement algorithms and integrate computing infrastructures. All you need to do is seleting which engine can meet your scenario. It's like you're in a restaurant. Take a look at the following menus.

Developers Delight

We obsessively seek out elegant, composable abstractions that enable robust, scalable, flexible integrations. Because we eliminate needless complexity and extraneous details, you can get up and running with Rosetta.ai in just a couple of minutes.

  • Simple integration
  • 30+ ready-to-use engines with unified APIs
  • Free trial to get started
                        
$request->setPostFields(array(
  'entity_id' => '100131585368796',
  'target_id' => '100131139421385',
  'event' => 'view'
));

try {
  $response = $request->send();

  echo $response->getBody();
} catch (HttpException $ex) {
  echo $ex;
}

                      
                        
require 'uri'
require 'net/http'

url = URI("https://api.rosetta.ai/event")

http = Net::HTTP.new(url.host, url.port)

request = Net::HTTP::Post.new(url)
request.body = "entity_id=100131585368796&target_id=100131139421385&event=view"

response = http.request(request)
puts response.read_body

                      
                        
import http.client

conn = http.client.HTTPConnection("api,rosetta,ai")

payload = "entity_id=100131585368796&target_id=100131139421385&event=view"

conn.request("POST", "event", payload, headers) # ignore headers here

res = conn.getresponse()
data = res.read()

print(data.decode("utf-8"))

                      
                        
var request = require("request");

var options = { method: 'POST',
  url: 'https://api.rosetta.ai/event',
  headers:
   { Authorization: 'Bearer API_TOKEN',
     Accept: 'application/vnd.rosetta-ai.v1+json',
     'Content-Type': 'application/x-www-form-urlencoded' },
  form:
   { entity_id: '100131585368796',
     target_id: '100131139421385',
     event: 'view' } };

request(options, function (error, response, body) {
  if (error) throw new Error(error);

  console.log(body);
});

                      
                        
package main

import (
  "fmt"
  "strings"
  "net/http"
  "io/ioutil"
)

func main() {

  url := "https://api.rosetta.ai/event"

  payload := strings.NewReader("entity_id=100131585368796&target_id=100131139421385&event=view")

  req, _ := http.NewRequest("POST", url, payload)

  res, _ := http.DefaultClient.Do(req)

  defer res.Body.Close()
  body, _ := ioutil.ReadAll(res.Body)

  fmt.Println(res)
  fmt.Println(string(body))
}

                      
                        
HttpResponse response = Unirest.post("https://api.rosetta.ai/event")
  .header("Content-Type", "application/x-www-form-urlencoded")
  .header("Accept", "application/vnd.rosetta-ai.v1+json")
  .header("Authorization", "Bearer API_TOKEN")
  .body("entity_id=100131585368796&target_id=100131139421385&event=view")
  .asString();

                      

Looking for enthusiasts to join our open source projects.  

Fast. Flexible. Future-proof.

Make recommendation engine simple for any scenario, anywhere.

Dynamically adjust and configure algorithms and computing resources based on your real-time data.

Developer First

Rosetta.ai is always-improving to provide toolkits, such as API clients, SDKs, embedded code and plugins with documentation to eliminate the chaos of developing recommendation. We're keeping working hard to make them production soon.

Plug and Play

We meticulously designed out-of-the-box recommendation engines for real-world practical scenarios. The secret sauce is our system dynamically adjust and configure necessary algorithms and resources via real-time data input.

Uncover Preference

Our system predicts and infers end user's personal preferences to help our customers understand their consumers, shoppers, or audience. We use data, process data and love data. We're your partner to uncover the latent iceberg beneath the sea.

Privacy Shield

Customers and end users are the boss and owner of their data. We commit and do our best to protect data privacy for individuals. Besides, we work for new infrastructure to fight for privacy protection. If interested to learn more, contact us anytime.

Customers Love Us

People love and trust Rosetta.ai. Don’t take our word for it.

Fall in love with Rosetta.ai it's crazy. I built my recommendation engines in few minutes. Just incredible! If you got money to spend, @rosetta_ai is great. I've used 20+ engines for my different scenarios and never been disappointed. Simply amazing!

Stunned by this product because of the variety and simplicity. A well constructed toolkit for people beginning to develop for recommendation features on product. I recommend @rosetta_ai to everyone!

People at Rosetta.ai really provide the best curated products out there. Looking for unique, on of a kind pieces, then this is the place to go. Every so often a new product comes out. At first you are hesitated, but then after trying it you're just in awe. Awesome!

Affordable Pricing

Our APIs are free to try and all come with a generous forever free plan. They’re fast and easy to set up, and if you need help, contact is just one click away.

  • Simple integration
  • 30+ ready-to-use engines with unified APIs
  • Free trial to get started

  • Records = # of Users + # of Items
  • Queries = # of Recommendation Request
  • QPS = Query per Second
  • Advanced Data Format: The Experiment data fields

Budget-limited Program

After 30 days trial of Essential or Business plan,
you can know your precise potential monthly fee and setup monthly budget, but not exceeding the plan's baseline price.

Applying for Budget-limited Program, the recommendation query only works in your peak time, which is measured by previous usage last month, but behavior tracking of user actions works all day long.

30 Days Trial and Budget-limited Program can help you to measure the ROI and the confidence to fully apply a paying plan.

Pioneer

Everyone should be able to build great recommendation.

Forever $Free /mo


  • 1 engine
  • 10K records
  • 100K queries
  • 30 QPS
  • Document Support

Essential

Scale at your own pace and pay for what you need.

Starting at $39 /mo (with discount for prepaid annual commitment)


  • 3 engines
    ( $5 / per additional 1 engine )
  • 50K records
    ( $15 / per additional 50K records )
  • 500K queries
    ( $10 / per additional 500K queries )
  • 75 QPS
  • Email Support
  •   Advanced Data Format
  •   Unlisted Scenarios
  •   30 Days Trial
  •   Budget-limited Program

Business

Increased features and limits to meet your needs.

Starting at $2500 /mo (with discount for prepaid annual commitment)


  •   Unlimited engines
    ( All scenarios enabled )
  • 2M records
    ( $300 / per additional 2M records )
  • 4M actions
    ( $200 / per additional 4M actions )
  • Custom QPS
  • Developer Support
  •   Advanced Data Format
  •   Unlisted Scenarios
  •   30 Days Trial
  •   Budget-limited Program

Automate your recommendation engine development now!