Serverless+Vector search

Amazon OpenSearch Service

Serverless : Auto-scaling Operation
Vector Search : Semantic search powered by Generative AI

My Role:
– Lead on the design of the end-to-end experience of the Serverless project from scratch. Such as security flow strategy, information architecture, and interaction design.
– Validated the design by conducting usability testing.
– Work closely with PM and engineer to power the serveless with generative AI.

Impact:
– Since the Serverless project release, daily actives have crossed 600 accounts and daily revenue has crossed $16K+, translating to $6M+.
– There is a large increase in the collections and indices created. Customers have created 3,500+ collections. One customer from GE Appliance shared that they are happy with the Serverless experience and will move larger workloads.

Teammates: 

1      x   UX Designer (Me)

1      x   Product Manager

1      x   UX writer

30+ x   Engineers (API team + Console team)              

Designed features for Serverless

Understand Context​

OpenSearch is a fully open source search and analytics engine. Use cases include: Real-time application monitoring, log analytics, and a full-text search application.

The auto-provisioning option allows customers to manage all aspects of their operations with minimal setup and maintenance.

A growing number of applications involving large language models are generating generative AI. Customers are increasingly interested in leveraging semantic search to provide better quality contextual search results.

How to drive user-centered design for complex configuration products?

Know your users

   #1 Know user case and consider data size for scalability

#2 Know users’ roles and collaboration behavior

How do those considerations affect design?

Translate and educate users

   # Hi-fi prototype 

Others

Support team on booth postcards, banners, hackathon phone tools, and clothing logos.