Content Recommendation

Orchestrates content recommendation through a multi-stage Conductor workflow.

Input: userId, contentType, maxResults | Timeout: 60s

Pipeline

crm_analyze_history
    │
crm_compute_similarity
    │
crm_rank_results
    │
crm_apply_filters
    │
crm_serve_recommendations

Workers

AnalyzeHistoryWorker (crm_analyze_history)

Reads viewedItems. Outputs viewedItems, likedItems, topCategories, activityScore.

ApplyFiltersWorker (crm_apply_filters)

Reads filteredItems. Outputs filteredItems, removedCount.

ComputeSimilarityWorker (crm_compute_similarity)

Reads similarItems. Outputs similarItems, id, score, category.

RankResultsWorker (crm_rank_results)

Reads rankedItems. Outputs rankedItems, rankingStrategy.

ServeRecommendationsWorker (crm_serve_recommendations)

Reads recommendations. Outputs recommendations, servedCount, responseTimeMs, trackingId.

Tests

4 tests cover valid inputs, boundary values, null handling, and error paths.

mvn test

---

Run this example: see RUNNING.md for setup, build, and CLI instructions.