Kieran Didi
Senior Machine Learning Scientist · Seattle, US
I’m a PhD student working with Michael Bronstein at the University of Oxford and David Baker at the Institute for Protein Design. I’m also a Senior Machine Learning Scientist in NVIDIA’s Fundamental Generative AI team, where I work on generative models for protein design.
Education
PhD in Computer Science, University of Oxford · 2024–2027
- Research in generative ML for protein design and optimization, supervised by Michael Bronstein.
- ELLIS PhD Scholar and EPSRC fully-funded scholarship
- Supervised Principles of Programming Languages, Concurrent Programming and CompBio
MPhil in Computational Biology, Cambridge University · 2022–2023 · Distinction
- Lectures include adv. graph ML, protein design, information theory, probabilistic ML
- Supervised algorithms course, designed and lectured course on SDEs+Generative Modelling
- 1-month research stay at MILA (Montreal, CA) as a visiting researcher for ML in Drug Discovery
Master of Science - MS in Biochemistry, focus Computer Science, Heidelberg University · 2021–2023 · 1.0 (best in year, German scale)
- Lectures from CS/Physics faculty, including machine learning I+II, GPU computing, operating systems, networks, software development, algorithms
- Lectures from Biology/Chemistry faculty, including bioinformatics and simulation methods
- Designed and lectured a course on data analysis with Python and bioinformatics
Exchange Scholarship in Chemistry (Part II), University of Cambridge · 2020–2021
- Natural Sciences Part II, lectures including math. methods, symmetry, theor. chemistry
Bachelor of Science in Biochemistry, Heidelberg University · 2018–2021 · 1.0 (best in year, German scale)
- Additional lectures in statistical learning
Work experience
Senior Machine Learning Scientist, NVIDIA · Remote + San Francisco, US · 2024-01–present
ML Research in GenAIR team under Arash Vahdat on generative modelling for biology
Deep Learning Researcher (Intern), Absci · NYC, US · 2023-09–2023-12
ML and data engineering for antibody design via geometric deep learning
Machine Learning Engineer Intern (part-time during studies), EMBL-EBI · Cambridge, UK · 2022-10–2023-06
Critically test and evaluate BioML models
Software Engineering Intern (Biosecurity, Cloud-native genomics), CSIRO Sydney, team Translational Bioinformatics · Sydney, Australia · 2022-07–2022-10
Developing cloud-native Python software for omics analysis
- Published open-source Terraform module for Spark on AWS (>3k downloads)
Consultant for ML/bioinformatics projects, PTNG Consulting · Melbourne, Australia · 2022-06–2024-01
ML for protein design/optimization and bioinformatics analysis
Thesis Student, Bernardes/Knowles/Sormanni Lab, Chemistry Department, Cambridge University · Cambridge, UK · 2020-09–2021-10
Developed peptide therapeutics and protein screening system (Nature Communications)
Research Intern, BioMed X Innovation Center (with Janssen Pharmaceuticals) · Heidelberg, Germany · 2019-08–2020-04
Established a screening system for autoimmune diseases for multi-national pharmaceutical company, enabling personalized screens
Research publications
Publications with author lists and project resources
- tmol: a GPU-accelerated implementation of the Rosetta energy function for machine learning. arXiv, 2026.
- Biomolecular design through latent reasoning with RFD4-Proteina. bioRxiv, 2026.
- Programmable amide bond formation using de novo ligases. bioRxiv, 2026.
- Adapting biological mechanisms for agentic protein design. preprint, 2026.
- Fast remote nucleotide sequence alignment with Riboseek. bioRxiv, 2026.
- Latent Generative Search unlocks de novo Design of Untapped Biomolecular Interactions at Scale. bioRxiv, 2026.
- AlphaFold Database expands to proteome-scale quaternary structures. bioRxiv, 2026.
- A universal model for drug-receptor interactions. bioRxiv, 2025.
- Accelerating Biomolecular Modeling with AtomWorks and RF3. bioRxiv, 2025.
- MotifBench: A standardized protein design benchmark for motif-scaffolding problems. arXiv, 2025.
- BioNeMo Framework: a modular, high-performance library for AI model development in drug discovery. arXiv, 2024.
- BioModelsML: Building a FAIR and reproducible collection of machine learning models in life sciences and medicine for easy reuse. bioRxiv, 2023.
- Online Search-Augmented Reinforcement Post-Training for All-Atom Protein Design. OpenReview, 2026.
- Multimodal Latent Flow Matching for De Novo Nanobody Design. OpenReview, 2026.
- Multi-Scenario Generative Protein Design via Preference Alignment. OpenReview, 2026.
- Proteina-Reforza: Reinforcement Learning Protein Binder Design. OpenReview, 2026.
- Proteina-Pronto: Fast De Novo Protein Design and Structure Prediction via Distillation. OpenReview, 2026.
- Towards All-Atom Protein World Models: Learning and Diagnosing Predictive Latent Representations. OpenReview, 2026.
- DMSsim: a statistical design framework for count-based sorting experiments that generate AI-ready sequence–function data. Manuscript in preparation.
- Atomic functional site scaffolding in biopolymers with RFdiffusion3. Science, 2026.
- De novo design of cysteine proteases. Nature, 2026.
- Computational design of metalloproteases. Nature, 2026.
- Efficient protein structure prediction from compact computers to datacenters with OpenFold-TRT. Computational and Structural Biotechnology Journal, 2026.
- Predicting PROTAC off-target effects via warhead involvement levels in drug–target interactions using graph attention neural networks. Computational and Structural Biotechnology Journal, 2025.
- GPU-accelerated homology search with MMseqs2. Nature Methods, 2025.
- Structure-based Drug Design with Equivariant Diffusion Models. Nature Computational Science, 2024.
- MISATO: machine learning dataset of protein–ligand complexes for structure-based drug discovery. Nature Computational Science, 2024.
- Synsor: a tool for alignment-free detection of engineered DNA sequences. Frontiers in Bioengineering and Biotechnology, 2024.
- AbNatiV: VQ-VAE-based assessment of antibody and nanobody nativeness for hit selection, humanisation, and engineering. Nature Machine Intelligence, 2023.
- Biomolecular condensate phase diagrams with a combinatorial microdroplet platform. Nature Communications, 2022.
- RNA-FrameFlow: Flow Matching for de novo 3D RNA Backbone Design. Transactions on Machine Learning Research, 2025.
- FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications. ICML (oral presentation), 2026.
- Scaling Atomistic Protein Binder Design with Generative Pretraining and Test-Time Compute. ICLR (oral presentation), 2026.
- La-Proteina: Atomistic Protein Generation via Partially Latent Flow Matching. ICLR, 2026.
- Compositional Flows for 3D Molecule and Synthesis Pathway Co-design. ICML, 2025.
- Proteina: Scaling Flow-based Protein Structure Generative Models. ICLR (oral presentation), 2025.
- DEFT: Efficient Finetuning of Conditional Diffusion Models by Learning the Generalised h-transform. NeurIPS, 2024.
- Dynamics-Informed Protein Design with Structure Conditioning. ICLR, 2024.
- Evaluating Representation Learning on the Protein Structure Universe. ICLR, 2024.
- Flows, straight but not so fast: Exploring the design space of Rectified Flows in Protein Design. ICML GenBio, 2025.
- Consistent Synthetic Sequences Unlock Structural Diversity in Fully Atomistic De Novo Protein Design. NeurIPS AI4Science, 2025.
- A framework for conditional diffusion modelling with applications in motif scaffolding for protein design. NeurIPS AI4D3 (oral presentation), 2023.
- Benchmarking Generated Poses: How Rational is Structure-based Drug Design with Generative Models?. NeurIPS MLSB (oral presentation), 2023.
- Modelling biology in novel ways - an AI-first course in Structural Bioinformatics. NeurIPS AI4Science, 2023.
- Flexible Small-Molecule Design and Optimization with Equivariant Diffusion Models. ICLR MLDD, 2023.
- Generating machine-learning-ready enzyme sequence-function data from microdroplet screening. Methods in Enzymology, vol. 736, 2026.
- Artificial Intelligence and Machine Learning in Bioinformatics. Encyclopedia of Bioinformatics and Computational Biology, 2nd edition, 2025.
Teaching
- Principles of Programming Languages, teaching, Oxford University, 10-12/2026
- Concurrent Programming, teaching, Oxford University, 10-12/2026
- Machine Learning Fundamentals, teaching, Oxford University, 08/2026
- Machine Learning for Science, 2-week lecture course, Oxford Summer School, 07/2026
- Bioinformatics, seminar series, Oxford University, 10/2025-02/2026
- Generative Modelling in Protein Design and Bioinformatics, guest lecture in bioinformatics course, Stanford University, 07/2025
- Deep Learning in Structure-Based Drug Design, tutorial, Spring School Structure-based Computer-aided Drug Design, Swiss Institute of Bioinformatics (SIB), 06/2024
- SDEs and Diffusion Processes, lecture, Cambridge University (MPhil Advanced Computer Science), 01/2024
- Structural Bioinformatics, lecture, Heidelberg University (BSc Biochemistry), 10/2023
- Graph Modelling with Applications to Structural Biology, presentation, CCAIM AI and Machine Learning Summer School, 09/2023
- Advanced Python for Scientists, lecture, Heidelberg University, 06/2023
- Algorithms II, supervising course on data structures and graph algorithms, Cambridge University, 03-05/2023
- Algorithms I, Cambridge University, 01-03/2023
- MLOps with PyTorch Lightning and W&B, workshop, ML Forum CSIRO Australia, 01/2023
- Python for Scientists, lecture, Heidelberg University, 11/2022
- Python Best Practices, workshop (co-host), Scientific Software Center Heidelberg, 11/2022
- Deep Learning with PyTorch, Lunch Time Python workshop, Scientific Software Center Heidelberg, 10/2022
- Python for Biochemists, lecture, Heidelberg University, 03/2022
Selected talks and presentations
- RFD4-Proteina and recent results, Google DeepMind, London (UK), 11/2026
- RFD4-Proteina for general protein design, Dayhoff Labs, London (UK), 11/2026
- Enzyme design and RFD4-Proteina, Ellison Institute, Oxford (UK), 11/2026
- RFD4-Proteina, Atomworks, and Annotated CIFs, RosettaCommons Workflows (online), 11/2026
- The Proteina Model Family, In Silico, London (UK), 11/2026
- RFD4-Proteina: Biomolecular Design through Latent Reasoning, Starkly Speaking Reading Group (online), 11/2026
- Scaling Atomistic Protein Binder Design with Generative Pretraining and Test-Time Compute,
- Open tools & models at the frontier of biological science, The Next Generation Symposium (AITHYRA), Vienna (Austria), 11/2026
- Biomolecular design through latent reasoning with RFD4-Proteina, European RosettaCon, Lisbon (Portugal), 10/2026
- What we can and cannot learn from general GenAI: Universal molecular design with Proteina models, BZH, Heidelberg University (Germany), 10/2026
- Latent generative search unlocks de novo design of untapped biomolecular interactions at scale, EMBL, Heidelberg (Germany), 10/2026
- From Pixels to Proteins: Scaling Generative AI for Science, Charité, Berlin (Germany), 10/2026
- State-of-the-art protein design with latent reasoning, GTC Berlin (Germany)
- Unlocking glycan binding and other polar interactions via generative modelling, London Protein Design Day, Imperial College (UK), 06/2026
- What comes after atomistic protein design? AI x Bio conference, Wellcome Sanger Institute Cambridge (UK), 06/2026
- Latent Generative Search unlocks de novo Design of Untapped Biomolecular Interactions at Scale, ML4ProteinEngineering Seminar (online), 04/2026
- Scaling Atomistic Protein Binder Design with Generative Pretraining and Test-Time Compute, Starkly Speaking Reading Group (online), 03/2026
- La-Proteina: Atomistic Protein Generation via Partially Latent Flow Matching, ML4ProteinEngineering Seminar (online), 08/2025
- Why equivariance is (not) so popular in Life Sciences (invited talk), ICML ML4LMS 2024, 07/2024
- A framework for conditional diffusion models with applications in motif scaffolding for protein design, ML4ProteinEngineering Seminar (online), 06/2024
- Conditioning generative models for proteins, images and more (oral presentation), AI and Biology Conference, EMBL Heidelberg (Germany), 03/2024
- Where do we stand in protein and ligand design? (oral presentation), Computational Structural Biology Conference, EMBL Heidelberg (Germany), 12/2023
- Conditional Diffusion Models for Protein Design via Doob’s h-transform (poster), MoML Conference, MIT (US), 11/2023
- Reality Check - Evaluating Generative Models for Protein and Ligand Design (poster), Machine Learning in Drug Discovery Symposium, Broad Institute (US), 10/2023
- Accessible AMR-detection tools for clinicians via cloud-based bioinformatics (oral presentation), Clinical Informatics Symposium, Melbourne (Australia), 12/2022
- sINSIDER: a cloud-native modular platform for kmer-based genomic analysis (oral presentation), ABACBS Conference, Melbourne (Australia), 11/2022
- Moving bioinformatics to the cloud (keynote), eSCAMPS Symposium, Cambridge (UK), 09/2022
- Detecting DNA integration via HPC bioinformatics pipeline (poster), R&I Conference, Sydney (Australia), 08/2022
Scholarships and awards
- 2025 · Duveen Scholarship St John’s College Oxford: ambassador for the college for a US trip
- 2025 · ELLIS PhD scholarship for interdisciplinary machine learning research
- 2024 · EPSRC DTP scholarship for academic excellence and full funding for Oxford PhD
- 2024 · Excellence certificate at Heidelberg University for best master’s degree in the year
- 2023 · Scholar’s award at St John’s College Cambridge for distinction in MPhil degree
- 2023 · Software Sustainability Fellowship: working on reproducibility in ML for Biology
- 2023 · Polaris Fellowship
- 2022 · AI Fellowship by Hummingbird VC
- 2022 · Kurt Hahn Scholarship at Cambridge University
- 2022 · Scholarship from both DAAD and Studienstiftung for master’s studies at Cambridge
- 2021 · Sartorius scholarship for master’s studies at Heidelberg
- 2021 · Marsilius Certificate for statistical learning: econometric predictions via LSTMs
- 2018 · Scholarship of German Academic Foundation based on intellectual ability
- 2018 · Biology Olympiad, 2nd place in Germany (>2,000 participants), silver medal at the International Competition in Tehran (best 30% of global selection of talents)
Voluntary experience
Founder & Co-lead, Computational Motifs Hacking Group · 2024-07–present
Organized monthly sessions on accelerating science with computation
Communications Lead, Cambridge Chapter, Nucleate UK · 2022-07–2024-07
Organized nationwide comms strategy for empowering biotech talents