Real-time ad ranking and creative review on a typed evaluation model — one call prices a whole auction, no trained CTR model, no logged clicks.
-
Updated
Sep 20, 2026 - JavaScript
Real-time ad ranking and creative review on a typed evaluation model — one call prices a whole auction, no trained CTR model, no logged clicks.
Privacy-aware multimodal ad targeting, recommendation, and placement platform using Python, ML, Parquet, DuckDB, DVC, and explainable ranking.
CTR prediction, candidate ad ranking, cold-start handling, drift monitoring, and low-latency ML serving for contextual ad systems.
Experimental study of pCTR prediction, calibration, value-aware ranking, and auction decision robustness using real RTB logs.
Educational exploration of how AI, ad ranking, auctions, and campaign metrics can influence product visibility in Mercado Libre Product Ads.
To associate your repository with the ad-ranking topic, visit your repo's landing page and select "manage topics."