Experience

  1. Computational Biologist

    Imperial College London - Ledesma-Amaro Synthetic Biology group

    Activities include:

    • Identifying methods to apply AI for the development of sustainable food proteins
    • Using in-silico directed evolution to improve the functionality of enzymes relevant for bioproduction
    • Optimising neural network models for bacterial promoter prediction using Keras and PyTorch
    • Design, development and implementation of WASP, a Python-based command-line pipeline for structure-based protein annotation
    • Network analysis, statistical testing, and enrichment for protein function inference
    • Use of REST APIs for web scraping
    • Communication of findings through regular presentations and reports
    • Participating in the selection, evaluation and interview process of new candidates for collaboration
    • Supervising and mentoring of students

  2. Lab Technician

    Maastricht Center for Systems Biology

    Lab activities include:

    • PCR and quantitative PCR assays
    • Molecular cloning, Gibson assembly and esiRNAs construction
    • Lab resources management
    • Preparation of weekly reports on experimental procedures

  3. Bioinformatician

    IRCSS Institute of Neurological Sciences

    Activities include:

    • Whole-Exome Sequencing data analysis for variant analysis
    • Evaluation of phenotype-driven tools for variant prioritisation
    • Python, R and Unix coding to streamline Next Generation Sequencing data analysis
    • Preparation of regular reports and presentations to communicate findings

Education

  1. MSc Systems Biology

    Maastricht University

    GPA: 8.5/10, cum laude

    Thesis: WASP: A new Pipeline for Functional Annotation of Proteins using AlphaFold Structural Models

    Courses included:

    • Systems Biology and Modelling Biosystems
    • Machine Learning and Multivariate Statistics
    • Dynamical Systems, Non-Linear Dynamics and Dynamic Game Theory

  2. BSc Genomics

    Alma Mater Studiorum - University of Bologna

    GPA: 110/110, cum laude

    Thesis: Phenotype-driven Variant Prioritisation Tools: Analysis of Whole Exome Sequencing in Patients with Hereditary Optic Neuropathy

    Courses included:

    • Bioinformatics and Omics Technologies
    • Programming, Statistics and Data Science
    • Molecular Biology and Genetics
Technical skills
Programming Languages
Python
R
Unix
MATLAB
Developer tools
Git & GitHub
HTML/CSS
REST APIs
Neo4j & Cypher
Awards
Generative AI with Diffusion Models
NVIDIA ∙ July 2024

In this workshop, I learned more about Generative AI and its applications in denoising diffusion models, which are a popular choice for text-to-image pipelines.
Learning Objectives:

  • Build a U-Net to generate images from pure noise
  • Improve the quality of generated images with the denoising diffusion process
  • Control the image output with context embeddings
  • Generate images from English text prompts using the Contrastive Language—Image Pretraining (CLIP) neural network
See certificate
Fundamentals of Accelerated Computing with CUDA C/C++
NVIDIA ∙ July 2024

In this workshop, I learned the fundamental tools and techniques for accelerating C/C++ applications on massively parallel GPUs using CUDA. I gained skills in writing and parallelizing code, optimising memory migration between the CPU and GPU, and applying these techniques to accelerate a CPU-only particle simulator for significant performance gains.
Learning Objectives:

  • Write code for execution on a GPU accelerator.
  • Express data and instruction-level parallelism in C/C++ using CUDA.
  • Optimise memory migration with CUDA-managed memory and asynchronous prefetching.
  • Use command-line and visual profilers to optimise performance.
  • Utilize concurrent streams for instruction-level parallelism.
  • Develop and refactor GPU-accelerated CUDA C/C++ applications using a profile-driven approach.
See certificate
Deep Learning with PyTorch : Build an AutoEncoder
Coursera ∙ May 2024
Learned to implement an autoencoder using PyTorch. An autoencoder is a type of neural network that learns to reproduce its input at the output layer. It consists of two main parts: the encoder, that compresses the input into a compact representation, and the decoder, which reconstructs the input from this compressed representation. In this project, I implemented an autoencoder to denoise handwritten digits.
See certificate
Neo4j Certified Professional
Neo4j ∙ November 2023

Neo4j is a graph database management system that uses graph structures — comprising nodes, edges, and properties — to represent and store data. It is designed for handling large-scale connected data and is renowned for its high performance and scalability.
Learning objectives:

  • Understanding graph theory, the fundamentals of graph databases, and their differences from traditional relational databases.
  • Neo4j’s architecture and optimised query processing with the Cypher language.
  • Best practices for designing and implementing graph data models to represent complex relationships and ensure data integrity.
See certificate