Jan 2026 - Present
AI Scientist: Botnar Institute of Immune Engineering (BIIE)
Design and train machine learning models to predict protein-protein interactions for the de novo design of antibodies and TCRs.
Implement and deploy robust ML pipelines on HPC and cloud infrastructure to enable large-scale virtual screening and generative protein design.
Collaborate with experimental scientists to validate computational predictions and refine designs with high-throughput sequence and structural data.
Apr 2021 - Jun 2025
PhD in Computational Biology: EPFL (Bitbol Lab)
Specialized on: Deep Learning, Generative AI, Protein design, Interpretability of large neural networks.
Used and trained large deep learning models, specifically Transformers and State Space Models, on biological sequences, for different learning tasks (e.g. MLM, CLM, discrete diffusion and classification) using various ML frameworks (e.g. Pytorch, Pytorch Lightning, Huggingface).
Worked with large biological datasets, both sequence-based (e.g. UniProt, OAS, InterPro) and structure-based (e.g. PDB, SabDab, DIPS, AFDB).
Published 5 papers on high impact scientific journals. Obtained multiple awards and contributed talks at international machine learning and structural biology conferences.
Jun 2024 - Aug 2024
AI Research Intern: Absci
Led the company's efforts on using sequence data to improve the in-house deep learning models for de novo antibody design with antigen/epitope conditioning and CDR sampling via sequence-diffusion.
Implemented data processing pipelines for multiple antibody sequence databases and finetuned pre-trained protein language models on internal proprietary datasets.
Oct 2020 - Oct 2021
Research Assistant: Venice Long Data
Processed and cleaned historical deliberations of the Venice Senate using Python and NLP tools (spaCy, NLTK, TF-IDF, BERT embeddings) to create structured textual datasets. Developed feature matrices capturing semantic relationships between documents to enable large-scale text analysis.
Built document similarity networks based on cosine similarity and applied hierarchical clustering to identify thematic clusters within archival records. These analyses contributed to the Venice Long Data project’s goal of uncovering long-term social and economic patterns from historical sources.
2021 - 2025
École Polytechnique Fédérale de Lausanne (EPFL): PhD in Computational and Quantitative Biology
Thesis: Revealing and exploiting coevolution through protein language models
Advisor: Anne-Florence Bitbol
2018 - 2020
University of Padova: MSc in Physics - GPA: 110/110 cum laude
Thesis: The effect of Delayed Dynamics on the Stability of Ecosystems
Advisors: Samir Suweis, Sandro Azaele
2015 - 2018
University of Padova: BSc in Physics - GPA: 109/110
Advisors: Cinzia Sada, Zamboni Riccardo
26 Jul 2024
Best paper award (ICML Workshop):
ProtMamba: a homology-aware but alignment-free protein state space model.
ICML Workshop on Accessible and Efficient Foundation Models for Biological Discovery. 41st International Conference of Machine Learning (ICML), Vienna, Austria.
22 Jul 2022
Contributed talk award (ICML Workshop):
Generative power of a protein language model trained on multiple sequence alignments.
ICML Workshop on Computational Biology. 39th International Conference of Machine Learning (ICML), Baltimore, USA.
RAG-ESM: Improving pretrained protein language models via sequence retrieval
Spotlight paper: ICLR MLGenX Workshop, Singapore (Apr 2025)
ProtMamba: a homology-aware but alignment-free protein state space model
Best paper award: ICML AccMLBio Workshop, Vienna (Jul 2024)
Poster: ICLR MLGenX and GEM Workshops, Vienna (May 2024)
Pairing interacting protein sequences using masked language modeling
Contributed talk: Natural Language Processing in Biology and Chemistry, Bern (Mar 2024)
Contributed talk: EMBO Workshop on Computational Structural Biology, Heidelberg (Dec 2023)
Poster: NeurIPS Workshop on Machine Learning in Structural Biology, New Orleans (Dec 2022)
Generative power of a protein language model trained on multiple sequence alignments
Contributed talk: ICML Workshop on Computational Biology, Baltimore (Jul 2022)
Contributed talk: 2nd Biology for Physics conference, Barcelona (Jul 2022)
Contributed talk: AI4Science Day at EPFL, Lausanne (Jun 2022)
Poster: Biological sequence variation. From statistical modeling to structure, function, and evolutionary dynamics, Cargese (Apr 2023)
Poster: International Course on Multiscale Integration in biological systems, Paris (Nov 2021)