1. Noise (India’s largest smartwatch brand) uses it to handle millions of launch-day visitors, steering the right SKUs to the right buyers in real time.
2. W for Woman personalizes across thousands of SKUs, tailoring product order by region and occasion.
How it works: track micro-interactions → infer intent → re-rank instantly at the edge.
Results so far: +15–25% product page clicks, +10–18% AOV lift, lower bounce.
Would love HN’s thoughts on where personalization should stop, and how to make it sharper without over-collecting data.