> This page is for Unified.

> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.merge.dev/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.merge.dev/_mcp/server.

# Train AI with ATS data

> How to ingest ATS data for AI model training and recruiting optimization using the Merge ATS API.

---

#### User authorizes ATS access

Your users authorize your application to access their ATS platform by going through an authorization flow using Merge Link.

#### Retrieve comprehensive ATS data

Utilize Merge's API endpoints, such as [**GET** /candidates](https://docs.merge.dev/ats/candidates/#candidates_list), [**GET** /interviews](https://docs.merge.dev/ats/interviews/#interviews_list), [**GET** /applications](https://docs.merge.dev/ats/applications/#applications_list), and [**GET** /offers](https://docs.merge.dev/ats/offers/) to retrieve comprehensive ATS data.

#### Clean and transform data

Clean the data by removing duplicates in your customer data, handling missing values, and filtering out irrelevant fields.
Transform data into a format suitable for AI training, possibly involving normalization, encoding, or feature engineering.

#### Validate model performance

Validate the model's performance using the testing set. Compute relevant metrics to gauge the model's effectiveness.

#### Monitor and retrain

Continuously monitor ATS data for updates, changes, and new information and implement a feedback loop for model retraining and improvement to adapt.

#### Surface insights to customers

Build a UI to provide your customers with valuable insights generated by the AI models, such as using candidate data to identify top talents efficiently or ways to optimize your recruiting process.

#### Get started in automatically pulling ATS data to train AI with Merge

Integrate today or talk to our sales team to learn how Merge unlocks hundreds of integrations in days - not years.

\


Get a demo **

\
\