I'm available for web development and software consulting. Let's build something great together!
Recent Experience
Founder & Full Stack Developer
Bakney srl, September 2022 - Present
I write code and iterate rapidly to bootstrap Bakney Sport, a SaaS CRM and sport management software for Sport Associations in Italy. Take a look at the Bakney Sport website.
Full Stack Developer
Redder, May 2021 - March 2024
At Redder, I led significant projects across various technologies. I migrated the PBXpress product from a Perl monolith to a REST API backend with a new SPA frontend using Svelte and svelte-spa-router. This involved developing custom Svelte components and creating a compiler for Perl Time::Period to JSON conversion. I engineered microservices, including a Python Slack bot for Covid Greenpass validation and a Node.js service for PDU camera streaming via WebSockets. Additionally, I contributed to Angular and Go projects, and extended the functionality of rust-lang/mdBook for SPA compilation (ISSUE 982). My work encompassed a wide range of technologies including Perl, Svelte, Python, Docker, Angular, Go, and Node.js, demonstrating versatility in both frontend and backend development.
Past Experience
Data Scientist
BlueTensor, September 2020 - May 2021
- I developed a search algorithm using NLP for the World Health Organization to improve assistive products search. I started by extracting and cleaning data from multiple words files. I then designed a graph-based search algorithm to find the most similar documents. On top of that I implemented a BERT-based similarity check feature for rule conflict evaluation for the moderators users to avoid writing duplicate rules. I used the following technologies: PostgreSQL, Python, Django, neo4j, bash scripting, PyTorch, Docker, Redis, HTML, JavaScript.
- In another project I optimized an algorithm and implemented UI for an industry 4.0 tire machine software using C# and Microsoft SQL Server.
Software Engineer (freelance)
BCademy, February 2019 - September 2019
- I built the first version of BNotary implementing Opentimestamps protocol for document timestamping.
- In another project I developed an ERC20-compatible token with a smart contract using Solidity in a private self-hosted Ethereum network. I then integrated the token in an existing React Native mobile application.
Software Engineer
Accenture (via Modis), August 2018 - August 2019
- I worked on Bank Intesa San Paolo's frontend team, developing features for mortgage management web applications. I used Java, HTML, CSS, AngularJS, and IBM Toolkit to build interfaces. The project involved integrating with COBOL team's data and other web services via IIB middleware and SOAP API.
- I worked with Oracle DB and PL/SQL procedures, optimizing database queries. I also used Adobe LiveCycle to generate PDF documents.
- In another project, I contributed to renewing the general document archive using C#, improving record organization and accessibility.
Education
Master's degree
Università degli Studi di Trento, 2019 - 2021
Computer Science, Data Science specialization
Bachelor's degree
Università degli Studi di Trento, 2015 - 2018
Computer Science, interdisciplinary path in Economics and Finance
Highschool
ITIS G.Marconi (Verona), 2010 - 2015
Information Technology & Telecommunication
Technical skills
- Svelte
- Javascript
- Typescript
- PostgreSQL
- Redis
- SQLite3
- Django
- Flask
- Celery
- Python
- Github Actions
- Docker
- Ansible
- Git
- Linux
- Ubuntu
- Debian
- HTML
- CSS
- Tailwind CSS
- LLM
- Genetic Programming
- Transformer Models
Other skills
- Google Ads
- Meta Ads
- Google Analytics
- SEO
- Software as a Service (SaaS)
- Product Management
Soft skills
- Accountability
- Initiative
- Problem Solving
- Active Listening
- Empathy
- Curiosity
- Creativity
Open Source & Projects
Publications
Genetic Improvement of TCP Congestion Avoidance
Conference paper @ BIOMA 2022, doi: https://doi.org/10.1007/978-3-031-21094-5_9, 3 citations
In this paper, we used Genetic Programming (GP) to evolve novel TCP congestion policies, encoded as C++ programs. The results show that the protocol discovered by GP is more aggressive in increasing the congestion window, leading to an improvement in network throughput of about 5%, showcasing the potential of GP for evolving novel congestion policies. I developed a custom GP framework to run the simulations in Python, and a wrapper for NS3 simulator fitness evaluation. Moreover, I built a mini web application in Svelte to configure, run, visualize, and export the simulation results. The code used in the paper is available in this GitHub repository.
More about me
I have always been passionate about technology and programming. I started learning Python at 13 years old through YouTube tutorials,
fascinated by the idea of creating computer programs.
I enjoy learning new things and believe that in programming, you can quickly adapt to new technologies as needed.
Crunching the numbers, I've been programming for about 15 years now, with more than 5 years of professional experience.