Profile navigating the latent space of life

email · github · scholar · linkedin · cv

about

I'm Dhruv Menon, an applied Machine Learning Scientist and Research Engineer building generative AI systems for scientific discovery. My work combines foundation models, chemical language models, representation learning, flow-based generation, and physics-aware simulation workflows to navigate chemical design spaces and accelerate scientific discovery.

I'm completing my PhD (thesis submission: Sep 2026) in Chemical Engineering at the University of Cambridge supervised by Prof. David Fairen-Jimenez. I build generative AI systems for the discovery & design of inorganic-organic materials called metal-organic frameworks for applications ranging from drug delivery to carbon capture & water harvesting. I also have interests in mechanistic underpinnings of performance, which I resolve using molecular simulations and statistical mechanics.

In 2025, I undertook a research fellowship at UC Berkeley with Prof. Omar Yaghi (Nobel Laureate, 2025), where I worked on generative materials discovery and automated laboratories. I earned my Master's by research (M.Res) in Physical Sciences (2023) from the Cavendish Laboratory, University of Cambridge and my Bachelors of Technology in Materials Sciences (2022) from the Indian Institute of Technology Gandhinagar.

Let's navigate the latent space together.

interests

I have eclectic research interests, however, the central `theme' that connects everything is computation, rational design and accelerated scientific discovery. My interests include (in no particular order): de novo molecular design, language models, diffusion models, flow models, small-molecule discovery, protein design, metal-organic frameworks & supramolecular materials, physisorptive carbon capture, gas storage & separations, and drug delivery.

On a philosophical note, I'm NOT interested in incremental research (atleast not anymore). I tend to focus on moonshots, and would like to surround myself with people with the same mission & ambition.

research

Ongoing & recent

Inverse design of metal-organic frameworks

Developed Nexerra, a generative molecular design framework combining transformer-based chemical language models, variational autoencoders, and conditional flow matching for inverse molecular design. Developed representation learning methods for molecular embeddings, latent-space optimization, and conditional molecular generation under chemical validity, synthesizability, and physical design constraints. Implemented objective steering for end-to-end AI-guided materials discovery for adsorption, gas storage, water harvesting, and drug delivery.

Predicting protein conformations

Developing all-atom equivariant neural networks to accelerate molecular dynamics simulations and improve sampling of protein conformational landscapes. Exploring trajectory-learning approaches and protein foundation-model embeddings for predicting and generating alternative protein conformations.

Computational drug delivery & nanomedicine

Developed interpretable machine learning models to predict the biocompatibility of drug delivery systems, enabling high-throughput screening of ~100,000 experimentally reported materials for nanomedicine applications. Built a hierarchical virtual screening workflow to identify optimal drug carriers for chemotherapeutics. Discovered PCN-222 as a promising drug delivery platform, which was subsequently synthesized and tested in mouse models of pancreatic cancer, where the optimized formulation demonstrated reduced tumour growth and metastatic spread.

Confined proton transport

Developed multiscale simulation workflows integrating Grand Canonical Monte Carlo and molecular dynamics to investigate molecular transport, and proton-conduction mechanisms in functional porous materials. Studied diffusion, radial distribution functions, hydrogen-bond networks, velocity correlations, spatial distributions, and transport behaviour. Integrated molecular simulations with structural and experimental evidence to identify the mechanisms governing transport under nanoconfinement.

papers

20+ peer-reviewed publications across materials chemistry, molecular simulation, and AI-guided materials design, including Nature Chemistry, Advanced Materials, Chem, Matter, and Chemical Society Reviews. 800+ citations; h-index 16. Full list on . ϕ denotes co-first authorship.

notes

experience

  1. Winton Cambridge - Berkeley Exchange Fellowship UC Berkeley

    Awarded a research fellowship at University of California, Berkeley under the supervision of Professor Omar Yaghi to work on the “generative design of metal-organic framework (MOF) water harvestors”.

  2. Summer Undergraduate Research Fellowship California Institute of Technology

    Awarded a research fellowship at California Institute of Technology under the guidance of Professor Austin Minnich to work on the “reduction of electronic noise in semiconductors.” [held virtually due to COVID-19]

education

  1. PhD in Chemical Engineering University of Cambridge

    Thesis: "Computational Design of Metal-Organic Frameworks". Publications in leading journals across materials chemistry, molecular simulation, and AI-guided materials design, including Nature Chemistry, Advanced Materials, Chem, Matter, and Chemical Society Reviews

  2. M.Res in Physical Sciences University of Cambridge

    EPSRC CDT in Nanoscience and Nanotechnology (NanoDTC). Lab rotations with Prof. Jeremy Baumberg on nanophotonics and Prof. Erwin Reisner on adaptive evolution for syngas fermenting bacteria.

  3. B.Tech in Materials Science Indian Institute of Technology Gandhinagar

    Institute Gold Medal; Director's Silver Medal; Gold Medal for Outstanding Research; Dean's List and Academic Excellence Scholarships. Member of the Institute football team.

Website theme inspired by John Gardner.