Experience

PopVax · Hyderabad · Jun 2026 – present

Machine Learning Scientist

Machine learning for cancer therapeutics, from target to designed molecule, working toward treatment tailored to the individual patient.

Peptide–MHC class I complex, PDB 3MRP

01

AI for cancer drug discovery

Machine learning applied to therapeutic discovery for solid tumours that have resisted everything else.

02

De novo design using deep learning

Frontier generative models that design molecules from scratch rather than searching a catalogue, with the pipelines around them for screening and for finding what is worth targeting.

03

Personalized cancer treatment

AI driven drug discovery and de novo protein design, working toward therapies tailored to the individual patient.

About the structure

The molecule above is a peptide–MHC class I complex, built from PDB 3MRP: an HLA-A*0201 heavy chain (chain A, 275 residues) and β2-microglobulin (chain B, 100 residues) drawn as cartoon ribbons, holding up a ten-residue tumour-antigen fragment (chain P) in ball-and-stick. It is the interface a T cell reads when it decides whether a cell is healthy, and the interface a cancer immunotherapy has to design against.

Ribbons are splined through the Cα trace with helices and β-strands taken from the deposited secondary structure records; waters are omitted. Geometry was tessellated offline, packed into a single glTF binary with three named nodes, and meshopt-compressed to about 265 KB, fetched only when the section comes into view.

Structure data: RCSB PDB 3MRP · 10.2210/pdb3mrp/pdb