I am a PhD student at the Molecular AI department in AstraZeneca, as well as at Chalmers University. I mostly work on deep learning for molecular generation and representation learning, but also have other research interests, including neurosymbolic learning, reinforcement learning and evolutionary algorithms.
My most significant (publically available) recent work has been on Chemformer - a pre-trained transformer for seq-to-seq computational chemistry tasks, and MolFlow - a flow-matching model for 3D molecular generation.
If you have any questions about my work or would just like to get in touch, feel free to email me - rossir [at] chalmers [dot] se