Software Engineer - Knowledge Systems

4w4 weeks ago

Exa

San Francisco, US · Full-time · $180,000 – $350,000

About this role

Exa is an applied AI lab building a search engine unlike the world has ever seen, powering search for Cursor, Cognition, HubSpot, and over 400,000 developers. Our ultimate goal is to build perfect search over all the world's information, far beyond Google.

As a Software Engineer on the Knowledge Systems team, you'll help build systems that understand what is true about the world. You'll work on extracting, connecting, retrieving, and reasoning over knowledge from the web and beyond so AI agents can answer questions with unprecedented precision and completeness.

You'll join a team building massive-scale ML systems that will define the way the new AI world consumes information. Backed by $350M from Lightspeed, Benchmark, and a16z, we design super high-performant vector databases and train state-of-the-art embedding models to process the entire web.

This role offers the opportunity to work on groundbreaking projects like building a system that can answer every factual question about every person in the world. If you want to build systems that will define the future of information retrieval, this is the place for you.

Requirements

Experience building and maintaining complex distributed backend systems
Ability to build data pipelines that process petabytes of documents, metadata, and unstructured data
Comfortable optimizing systems for correctness, reliability, and scale
Deep care about data quality and finding high-quality information
Experience building systems involving graph traversal, entity resolution, data quality, or truth discovery (preferred)

Responsibilities

Build and maintain complex distributed backend systems for knowledge extraction and retrieval
Build data pipelines that process petabytes of documents, metadata, and unstructured data
Design systems for extracting, connecting, retrieving, and reasoning over knowledge from the web and beyond
Optimize systems for correctness, reliability, and scale
Build a system that can answer every factual question about every person in the world
Design a new query language for searching, traversing, and reasoning over billions of facts

Benefits

Opportunity to work on massive-scale ML systems that define the way the new AI world consumes information
Work with cutting-edge infrastructure to crawl the entire web and train state-of-the-art embedding models
Join a well-funded team backed by $350M from Lightspeed, Benchmark, and a16z
Contribute to a product that powers search for Cursor, Cognition, and HubSpot