CV

I’m a postdoctoral fellow at the University of Waterloo working at the intersection of AI and science. Previosuly, I worked in AI for Biology. I was a ML engineer at Gandeeva Therapeutics, where I designed proteins and antibodies. For my Ph.D., I designed representations for Biology, working with Libbrecht lab.

Work experience

• Postdoctoral Fellow, University of Waterloo (Oct 2023 - Present)

  • Consulting on multiple projects in AI for science.
  • NSERC fellow starting October ‘24

• Researcher, Royal Bank of Canada (July 2024 - Present)

  • Working on LLM’s + evolutionary strategies and Reinforcement learning from verifiable rewards (RLVR) for spatial optimization using heuristics.

• AI Consultant, Stealth Startups (Jun 2023 - Sep 2023)

  • Worked alongside the team to improving factual accuracy in LLMs. See thoughts compiled later here.

• Machine Learning Engineer, Gandeeva Therapeutics (Jun 2022 - May 2023)

  • Integrated molecular dynamics based conformation search with deep learning based protein design for protein affinity maturation. Gandeeva will use this platform for their future protein affinity maturation campaigns.
  • Designed antibodies using sequence and structure based Machine Learning, energy metrics, and bayesian optimization.

• Ph.D. Candidate, The University of British Columbia (Jan 2019 - Feb 2023)

  • Designed representation learning strategies for chromatin structure and the epigenome.

• Machine Learning Intern, Skycope Technologies (May 2018 - Sep 2018)

  • Built Skycope’s data and machine learning infrastructure. Integrated ML into Skycope’s existing software stack, which is now it’s flagship drone detection software.

Education

• Ph.D. in Electrical and Computer Engineering (Thesis: Representation Learning Strategies for the Epigenome and Chromatin Structure using Recurrent Neural Models), The University of British Columbia, Vancouver, 2023
• M.A.Sc. in Electrical and Computer Engineering, The University of British Columbia, Vancouver, 2018
• B.Tech. in Electronics and Communication Enginnering, National Institute of Technology Karnataka, India, 2017

Skills

• Machine Learning, Artifical Intelligence, Climate Change, Computational Biology

Publications and Patents

Structuring Collective Action with LLM-Guided Evolution: From Ill-Structured Problems to Executable Heuristics

Dsouza, K. B., Watt, G. A., Leonenko, Y., & Moreno-Cruz, J. (2025). Structuring Collective Action with LLM-Guided Evolution: From Ill-Structured Problems to Executable Heuristics. arXiv preprint arXiv:2509.20412.

BoreaRL: A Multi-Objective Reinforcement Learning Environment for Climate-Adaptive Boreal Forest Management

Dsouza, K. B., Ofosu, E., Amaogu, D. C., Pigeon, J., Boudreault, R., Maghoul, P., ... & Leonenko, Y. (2025). BoreaRL: A Multi-Objective Reinforcement Learning Environment for Climate-Adaptive Boreal Forest Management. arXiv preprint arXiv:2509.19846.

Bridging Farm Economics and Landscape Ecology for Global Sustainability through Hierarchical and Bayesian Optimization

Dsouza, K. B., Watt, G. A., Leonenko, Y., & Moreno-Cruz, J. (2025). Bridging Farm Economics and Landscape Ecology for Global Sustainability through Hierarchical and Bayesian Optimization. arXiv preprint arXiv:2508.06386.

Assessing the climate benefits of afforestation in the Canadian Northern Boreal and Southern Arctic

Dsouza, K. B., Ofosu, E., Salkeld, J., Boudreault, R., Moreno-Cruz, J., & Leonenko, Y. (2025). Assessing the climate benefits of afforestation in the Canadian Northern Boreal and Southern Arctic. Nature Communications, 16(1), 1964.

Learning representations of chromatin contacts using a recurrent neural network identifies genomic drivers of conformation

Dsouza, K. B., Maslova, A., Al-Jibury, E., Merkenschlager, M., Bhargava, V. K., & Libbrecht, M. W. (2022). Learning representations of chromatin contacts using a recurrent neural network identifies genomic drivers of conformation. Nature Communications, 13(1), 1-19.

Wireless threat detection device, system, and methods to detect signals in wideband RF systems and localize related time and frequency information based on deep learning

Koppisetti, N. R. S. V. P., Dsouza, K. B., Boostanimehr, H., & Mallick, S. (2022). U.S. Patent Application No. 17/825,304.

Latent representation of the human pan-celltype epigenome through a deep recurrent neural network

Dsouza, K. B., Li, A. Y., Bhargava, V., & Libbrecht, M. W. (2021). Latent representation of the human pan-celltype epigenome through a deep recurrent neural network. IEEE/ACM Transactions on Computational Biology and Bioinformatics.

A Downscaled Faster-RCNN Framework for Signal Detection and Time-Frequency Localization in Wideband RF Systems

Prasad, K. S. V., D’souza, K. B., & Bhargava, V. K. (2020). A downscaled faster-RCNN framework for signal detection and time-frequency localization in wideband RF systems. IEEE Transactions on Wireless Communications, 19(7), 4847-4862.

Conferences and Talks

Representation learning for biology

December 10, 2021

Talks at 4DN, RECOMB, MLCB, Epigenetics Meeting, Online

Teaching

CPEN 491: ECE final year undergraduate Capstone design project

Service

• Joint-Secretary at Autism Society of Udupi (ASU), a non-profit organisation in Udupi, India, that aims to create awareness about Autism among parents, teachers, health professionals, students, general public, and all the stakeholders, so that early diagnosis and early intervention could give the child maximum benefits

• Past mentor at Climate Hub, iGEM, UBC, Lets Talk Science, and Geneskool, Genome BC

Projects

Representation Learning Strategies for the Epigenome and Chromatin Structure using Recurrent Neural Models

Private machine learning: Extension to the bayesian setting and higher dimensional querying

Character Based Language Models Through Variational Sentence and Word Embeddings

Hybrid Precoding for mmWave Massive MIMO OFDM