Doubleword

A London-based company building infrastructure that makes large-scale AI inference much cheaper.

About Doubleword

Doubleword builds the infrastructure behind large-scale AI inference. Its product is an asynchronous inference cloud.

Its About page says the team is interested in what becomes possible when inference is 100x cheaper and is working to make that real.

The About page describes it as London-based. Job pages list engineering roles in inference systems.

It describes its people as researchers and systems engineers who have built their own inference engines, published research and optimised GPU code. This is its own description.

The roles we found are senior, research or member-of-technical-staff positions, and we found no graduate or internship programme.

What they do

• {'desc': 'Cheaper batch AI processing.', 'title': 'Async inference cloud'}
• {'desc': 'Fast engines for large models.', 'title': 'Inference systems'}

Key facts

• London-based (About page)
• Hiring inference systems engineers
• Team of researchers and systems engineers (own description)

Website: www.doubleword.ai