Giulio Carducci

Work Experience

June 2022 - Present

DATA ENGINEER

Revelio Labs

Revelio Labs aggregates millions of public employment records in a single place, creating a universal HR database, allowing to understand the workforce dynamics and trends of the market. I work as a data enginner in the ingestion side of things, managing numerous different data pipelines that ingest the data right into the database. I also manage cleaning and standardization of input records, and integration with the existing data.
- I redesigned the most valuable ingestion process of the company, making it more effient, robust and lightweight. I deployed it so that it runs daily in a fully automatic fashion.
- I contribute to several processes that ingest different types of records into the main database, adding new features, maintaning them over time and making sure that they are consistent.
- I am responsible for the daily flow of millions of records/terabytes of data.
- I managed the migration of the whole knowledge base to a new data warehouse, by writing a python package that clones the tables to their designed destination, while maintaining the defined schema and constraints. The package has been used by dozens of coworkers and hundreds of automatic processes.
May 2018 - May 2022

R&D SOFTWARE ENGINEER

Synapta

I started working at Synapta right after my Master's Degree, and I carried on the collaboration in parallel with the activity of the reasearch scholarship. I worked on data analysis and software development for artificial intelligence, and on the backend-side integration of such applications. I manly worked for the website contrattipubblici.org, but I have also worked on other minor projects.
- I applied several data analysis, natural language processing (NLP) and machine learning techniques to derive new data from the existing knowledge base
- I identified and classified references to the norm UNI-EN ISO 9001 inside public administrations' documents using information extraction and machine learning techniques.
- I devised machine learning and NLP strategies for out-of-core computation on huge data volumes
- Main backend web developer and QA analyst of contrattipubblici.org.
- I configured and monitored the company's main services on dedicated servers, according to data redundancy and high availability principles
- I managed the entire development process of about 250000 document of interest (search with APIs, download and insertion to database, Optical Character Recognition (OCR), classification and indexing in a search engine)
- I renewed the project TellMeFirst by rewriting it from scratch in python in a clearer, lighther and more modular way, and by making it available as an API.
- I worked on the app for Android devices BandiPubblici, in particular I took care of the development of its backend and the integration with the existing services, the release on the play store, and in-app purchases.
- I developed the entire backed side of the website Data for Citizen Engagement for the Fondazione Compagnia di San Paolo, including the computation of values for each graph starting from the raw data.
- I worked on the Wiki Loves Monuments project by developing a Authorization insertion form, for Wikimedia Italy.
June 2018 - June 2019

RESEARCH FELLOW FOR THE REPOSUM PROJECT

Università degli Studi di Torino & Synapta

The scholarship is funded by the Department of Philosophy and Educational Sciences of UniTo, and involves the annotation alaysis and representation of a corpus of metadata about italian and foreign PhD dissertations using natural language processing and machine learning techniques. I closely cooperated with the members of the Department of Philosophy and Computer Science to achieve the research goals.
- I presented the results of the research work of the team at the IRCDL conference in Pisa in 2019.
- Classification of Philosophy theses using hybrid machine learning approaches
- Determination of the subject of the thesis with a strategy based on topic modeling
- Feature extraction from documents with several techniques such as bag-of-words, entity recognition, entity linking
- Study and application of several graph mining techniques in order to detect similarity links between theses based on semantic features
- I have rewritten in python the tool RelFinder to use it programmatically in the research activity
Read the final report of the research activity (only available in Italian).
April 2015 - September 2015

RESEARCH INTERNSHIP ON HUMAN ACTIVITY INTERPRETATION

Cognitive Systems Group (Universität Bremen)

Cognitive Systems Group (CoSy) is a research center of the University of Bremen that studies how the human mind processes spatial information and the surrounding events, and develops tool that helps the humans in these tasks. I worked for 6 months as an intern researcher thanks to the Erasmus+ program, where I had the opportunity to collaborate with a multicultural and multidisciplinary team in the context of a european project (Cognitive Vision).
- Camera movement estimation based on detected motion in scene
- Applicazione of different methods for face detection and face recognition in the movie domain
- Training of custom shapes and objects detectors

Education

September 2015 - April 2018

MASTER DEGREE IN COMPUTER ENGINEERING

Politecnico di Torino

I chose the Software path, which is taught in english; I also included some courses from the Data Science path. I wored on my Thesis in the Istituto Superiore Mario Boella (ISMB), a research center that is highly focused in innovation. My work consisted in determining the personality of Twitter users based on their tweets using machine learning and natural language processing techniques.
- Thesis: Semantic Analysis to Compute Personality Traits From Social Media Posts
- Read the Thesis at this link (TLDR; take a look at the presentation)
- Final Grade: 99/110
September 2011 - March 2015

BACHELOR DEGREE IN COMPUTER ENGINEERING

Università degli Studi Dell'Aquila

I worked on my Thesis in the DEWS research center, with the objective of developing and testing a local database for an embedded face recognition system.
- Thesis: Design, implementation and exploitation of a Database to support an embedded face recognition system
- Final Grade: 102/110
September 2006 - July 2011

HIGH SCHOOL DIPLOMA

Liceo Scientifico "A. Bafile", L'Aquila

Liceo Scientifico Sperimentale PNI (Computer Science National Plan).
- Final Grade: 93/100

Publications

Carducci, G., Rizzo, G., Monti, D., Palumbo, E., Morisio, M., TwitPersonality: Computing Personality Traits from Tweets using Word Embeddings and Supervised Learning, Multidisciplinary Digital Publishing Institute (MDPI), 2018. doi.org/10.3390/info9050127
Carducci G., Leontino M. Radicioni, D. P. Bonino, G., Pasini E., Tripodi P., Semantically Aware Text Categorisation for Metadata Annotation, Italian Research Conference on Digital Libraries (IRCDL), 2018.