Inside ConvergeAB: from one prompt to a better antibody

In this 60-minute session, Converge Bio Co-founder and CSO Iddo Wiener opens up ConvergeAB, our platform for antibody optimization. Prompt in an antibody and its target, and ConvergeAB returns a ranked list of lab-ready candidates, with no task-specific training and no human in the loop.

Iddo walks through how the platform was built and why its zero-shot approach removes the per-program cold start that slows conventional optimization. He then turns to the evidence: a zero-shot cetuximab biobetter that beat the clinical antibody on affinity, and a four-target campaign spanning PD-L1, SARS-CoV-2 Spike, HER2, and CD19 that found a stronger binder on every target, with the CD19 result mirroring an FDA-approved therapeutic.

True to a technical session, he is also candid about the limits: the properties beyond affinity that a therapeutic must clear, and the open question of how far the model's generalization extends.

Watch the full recording to see how ConvergeAB works, what it has delivered, and where the science goes next.

Who should watch: antibody engineers, computational and translational scientists, and R&D leaders evaluating AI-driven optimization.

Shorter version, if the page needs a compact blurb:

Converge Bio Co-founder and CSO Iddo Wiener takes you inside ConvergeAB, our antibody optimization platform: one prompt in, a ranked list of lab-ready candidates out, no task-specific training. He walks through the zero-shot cetuximab biobetter that beat the clinical antibody on affinity, a four-target campaign that improved on every target, and the open questions the platform is still working through. 60 minutes, presented for a technical audience.