Integrating AI-Powered Code Review into CI Pipelines with JD's Large Model

Manual code review suffers from high time consumption, inconsistency, missed errors, and subjectivity, which hampers development efficiency and code quality.

Leveraging AI large‑model services (e.g., JD's Chatrhino) can automate review, provide consistent feedback, and reduce human error, while still allowing human‑AI collaboration.

Benefits of AI‑driven review:

Significant time savings and faster development cycles.

Improved accuracy through continuous learning of the model.

Ensured code consistency with best‑practice enforcement.

Instant feedback for developers.

Knowledge sharing via AI‑generated suggestions.

Integration steps:

Connect to JD's large‑model API (Chatrhino) and obtain an API key.

Implement an AI review script (Java/JUnit5 example) that fetches the merge request, obtains the code diff via the Coding API, sends the diff to the model, and posts the review result back as a note.

Configure a CI pipeline (e.g., MCube) to trigger the script on merge‑request creation or update, including necessary atoms for code checkout, Maven build, and notification.

Set up webhook integration in Coding to automatically create merge requests on push events.

Example Java script (place in your repository):

void aiCodeReviewByChatrhino() {
//1、get mr
String mergeRequest = System.getProperty("mergeRequest");
if (mergeRequest == null) return;
System.out.println("【流水线触发MergeRequest】------------------------------------------------ " + mergeRequest);
//2、get coding diff
String codeDiff = getCodingDiff(mergeRequest);
System.out.println("【coding的diff内容】------------------------------------------------" + codeDiff);
//3、gpt code review
String result = gptReview(codeDiff);
System.out.println("【京东言犀大模型评论】------------------------------------------------" + result);
//4、output review content
note(objectMapper.readTree(result).get("choices").get(0).get("message").get("content").asText(), mergeRequest);
System.out.println("【coding评审note记录】------------------------------------------------" + noteContent + "-----------------------------------------------");
}

The pipeline then runs the script with Maven, for example:

mvn test -Dtest=com.jdwl.wms.common.AiCodeReview -DmergeRequest=${globalParams.user.WEBHOOK_ATTR_MERGE_REQUEST_ID} -DfailIfNoTests=false -Dmaven.test.failure.ignore=true

After deployment, metrics showed a reduction of development‑stage effort from 62% to 52% (≈10% drop), a decrease in average requirement delivery time from 26.57 days to 17.14 days, and a drop in bugs per person from 14 to 6.

In summary, embedding AI code review into CI pipelines automates quality checks, accelerates delivery, and frees developers to focus on core innovation.