From Spam Fighting at Facebook to Vector Search at Rockset: How to Build Real-Time Machine Learning at Scale

Rockset extends its real-time analytics database to support machine learning with new vector search capabilities.

In this tech talk, Rockset VP of Engineering Louis Brandy shares his experience building spam fighting systems, including Sigma at Facebook, which had a high reliance on real-time data for machine learning.

Louis is joined by Senior Product Manager John Solitario who explains how Rockset fits into the larger data stack for real-time machine learning, the latest support for vector search and how customers are using Rockset ML capabilities in production.

We’ll cover how Rockset optimizes for real-time machine learning in the following ways:

Real-time streaming data for machine learning
Turn events into real-time features: Transform and index streaming data to generate real-time features
Fast search: Use Rockset’s Converged Index with metadata filtering to deliver fast, efficient results
Vector search capabilities: Support for distance functions including dot product, cosine similarity and Euclidean distance for known nearest neighbor (KNN) search.

Louis Brandy

Louis Brandy is the Vice President of Engineering at Rockset. Prior to Rockset, Louis was Director of Engineering at Facebook. During his time there, he was an early engineer and manager in Facebook’s Site Integrity organization where his team built much of the anti-abuse infrastructure that powers Facebook’s spam fighting, fraud detection, and other online, real-time classification systems.

He also worked on Facebook's RPC and service discovery ecosystem and built and supported the C++ infrastructure teams responsible for the overall health of the Facebook C++ codebase, working on compilers, sanitizers, linters, and core (and open-source) libraries like folly, jemalloc, and fbthrift.

John Solitario

John Solitario is a senior product manager at Rockset and drives ingest and machine learning product development at Rockset, which includes integration support, ingest transformations, collection management, and vector search. Before joining Rockset, John spent several years at Salesforce where he worked across a multitude of products including Mulesoft, Commerce Cloud, and Service Cloud. He received a B.S in engineering and an M.S. in computer science with a focus in artificial intelligence from Stanford University.