The next companies coming out of UK science: meet Pillar VC’s second Encode Cohort

The second cohort of Encode Fellows is testing a new pathway between frontier AI and UK science.

Many AI researchers have the technical ability to tackle difficult scientific problems, but lack access to the laboratories, specialist data and collaborators needed to do so. At the same time, UK R&D labs often struggle to access advanced AI and machine-learning expertise through conventional academic or commercial routes.

The Encode: AI for Science Fellowship, powered by ARIA and delivered by Pillar VC, is designed to close that gap.

From 1,429 applications across 116 countries, ten researchers have been selected for its second cohort, with four relocating to the UK to build here. Over a year, they’ll work with UK scientists and ARIA Creators, access compute and pursue ambitious projects across ARIA’s opportunity spaces.

“Encode is about giving exceptional AI talent the freedom to build solutions to scientific problems that really matter — particularly the frontier challenges ARIA is exploring through our opportunity spaces.” Pippy James, Deputy CEO, ARIA

Encode sits between academia and company building. Fellows have the freedom to explore ideas whose eventual output might be a research tool, open-source model, dataset, venture or new organisation. The first cohort has already produced papers, open-source tools, collaborations and two ventures.

Not every project will work as intended. Models may fail to generalise and predictions may not survive experimental testing. But those outcomes can still produce useful evidence, methods and new research directions.

Through Encode, we’re testing what becomes possible when exceptional AI talent, UK scientific expertise and the freedom to build are brought together.

‍

Read the full announcement on ARIA's website