agents

Today in AI for life science: agent benchmarks converge on a lesson we felt firsthand — the system around the model matters as much as the model — virtual-cell ‘world models’ arrive with a push for reproducibility standards, and the lab runs its first live agent-driven experiment.

Today in AI for life science: a benchmark shows AI ‘scientists’ still struggle to even find the right papers, agents arrive in ImageJ with reproducibility built in, and the Human Cell Atlas convenes as the retina gets mapped.

Today in AI for life science: Anthropic bets on agent infrastructure over bigger models, the virtual cell meets honest benchmarks, self-driving labs go mainstream, and spatial omics gets its foundation models — plus a new chapter from the lab.

Our BioEngine preprint is out on bioRxiv. It turns the explosion of community AI models into something a biologist — or an AI agent — can actually run, on a laptop or an HPC cluster, no code required.

Can genuine scientific discovery emerge from radically simple AI agent loops? We explore autonomous research agents built from minimal ingredients.

An agent-first platform that deploys bioimage AI models as autoscaling, agent-readable interfaces on cloud or HPC — and powers the “test run” feature on the BioImage Model Zoo. Described in a 2026 preprint.

Open-source framework that connects AI models, data, and instruments into shared workspaces — the backbone of the lab’s platforms.

An AI assistant that helps scientists find the right imaging technologies and facilities across the Euro-BioImaging network from plain-language questions.

Control microscopes in plain language — LLM agents that turn scientific intent into imaging, segmentation, and analysis.

A community-driven, AI-powered assistant that helps researchers navigate bioimaging tools, databases and services — and autonomously runs analysis tasks from plain-language prompts. Published in Nature Methods.