Harshit Agrawal

Title of the Talk:
AI-Driven Marketplace Monetisation: From Static Rules to Explainable Decisions at Global Scale — A Neuro-Symbolic Framework for Fees, Promotions, and Seller Incentives

Abstract :
Large-scale online marketplaces rely on complex rule-based systems to determine seller fees, promotions, advertising rewards, and other economic incentives. While deterministic rules provide control and compliance, they struggle to optimize decisions across millions of sellers with diverse behaviors and economic contexts.

This talk presents a neuro-symbolic AI framework that combines marketplace policy rules with predictive and causal machine-learning models to estimate the incremental impact of alternative monetization actions. A constrained AI decision layer selects the action that maximises marketplace value while respecting eligibility, budget, trust, and policy requirements.

The framework also produces human-understandable explanations describing why an action was selected and why alternatives were rejected. Through generalised marketplace scenarios, the talk explores how such systems can evolve from static rule execution toward adaptive, explainable, and trustworthy AI-driven monetisation at global scale.

Bio:
Harshit Agarwal is a Staff Engineer with 14+ years of experience designing and scaling distributed systems across global commerce, cloud, and telecommunications. At eBay, he serves as a technical lead for the Plutus monetization platform, supporting 112 million transactions daily and 3.18 million sellers. His expertise spans backend architecture, pricing and promotions, AI-accelerated engineering, and large-scale system integration, with multiple innovation awards and proven business impact.