Yogi Cahyo Nyono

Diagnosed inefficiencies in the legacy discount logic and engineered a personalised product recommendation model (XGBoost) to allocate subsidies. The model targeted high-relevance products for new users, outperforming the previous system and delivering a 4.49% uplift in New User Activation in the Philippines. Automated campaign workflow for product curation, reducing manual workload and accelerating time-to-market for promotions, contributing to a 21.99pp GMV uplift over prior campaigns. Established a daily monitoring system for Subsidy-to-GMV efficiency across five SEA markets, improving ROI visibility and enabling agile campaign adjustments. Led the product instrumentation strategy for new Search features, defining event taxonomy and success metrics, with funnel observability that reduced hot-fix response times from days to hours.

Engineered a unified Single Customer View (CDP) for 50M+ users by integrating identity resolution and LLM-based demographic prediction. The infrastructure became the backbone for hyper-personalised marketing, directly unlocking higher ROI for campaigns across the ecosystem.

Engineered a volatility prediction framework leveraging alternative ESG data (MarketPsych) to enhance risk modelling for FTSE 100 assets. The Random Forest model achieved a 12-20% RMSE reduction, earning a feature in a Refinitiv corporate blog for its applied value.