W O R K   E X P E R I E N C E
JAN 2017 – CURRENT
ED7 London, Indipendent data analytics consultant – London (UK)
Since 2017, I work as independent senior data consultant,
carrying out data projects in London. The specific kind of
activities strictly depends on the customer needs.
Main customers and achievements:
Trainline. Trainline is the biggest digital rail and coach platform in
the UK, I have been working as team leader, to deliver business intelligence projects using Pyspark, Google Cloud and AWS. Managed 3 people. Reporting to the head of data and to the permanent project leader UK Government. Helping the UK government to build data pipelines in order to score the prison escape risk (minister of Justice, Westminster). Python, R, AWS.
Expedia Group. Expedia is the biggest travel agency in the
world.Led a project to tune their data infrastructure as well as
building machine learning algorithms for their marketing campaigns.
Gazprom. Gazprom is the biggest russian oil and gas company. I worked in the algo-trading group to help traders to productionalise their trading strategies for the oil and gas markets. Build a machine learning as a service tool available to all the traders through rest-api.
HSBC. HSBC is the biggest bank in Europe. Helping the trading
team to design and develop data pipelines for fair pricing Foreign
exchange rates. Python, Spark, Sklearn.

JAN 2016 – JAN 2017
Stratagem (Blockchain), Algo trader engineer – London, UK
Stratagem was a fintech startup, specialised in Financial
markets trading and sport betting. It has been acquired by
Blockchain in 2019.
Key tasks:
Parallelise back-testing platform using Apache Spark to
explore the input space for trading strategies(Genetic Algorithm, Reinforcement learning).
Developing of high-performance trading strategies able to
exploit arbitrage opportunities sometimes available in the markets
Tech Stack: Python, Cython, Cassandra, MongoDB,
Spark,Java, Rest API, Golang, Amazon S3, Kafka

JUNE 2014 – JAN 2016
Data Reply , Big Data consultant, Milan, Italy
Data Reply is an Italian data consultancy active in Italy,
Germany and UK. I carried out several projects, mainly
using Cloudera as Hadoop ecosystem. The most interesting project
was about building a customer classification system, scoring gender/age and purchasing power.

S K I L L S
Big data Spark, Hive, Hadoop, Redshift, BigQuery, Athena
Cloud Amazon (AWS), Google Cloud
AI /ML/ stats: Neural network, SVM, clustering, decision tree,
time series analysis, LSTM, hypothesis testing,
Languages: Python as main one, C and C++ when I have to,
Java if there is a JVM constraint. Golang for microservices.
DevOps: Docker, Jenkins, Terraform.
Methodologies: Scrum, Agile

P E R S O N A L   P R O F I L E
Extremely motivated to
constantly develop my skills
and grow professionally. I am
confident in my ability to
deliver massive data
analytics projects.

E D U C A T I O N
University of Pisa
Master degree in Computer engineering (Ingegneria Informatica), 110 e lode (cum laude). 2014. Dissertation
completed at Queen Mary University of London.

CERTIFICATIONS
Recurrent Neural Network.
Deeplearning.ai 2020
Programming in GO
(Specialization on Coursera,
University of California, Irvine)
2020
Time Series Analysis, University of
New York on Coursera, 2019
Sentiment Analysis, University of
Mitchigan, 2019
Improving Deep Neural
Networks: Hyperparameter
tuning, Regularization and
Optimization, deeplearning.ai,
2018
Country Level Economics:
Macroeconomic Variables and
Markets, University of Illinois,
Cousera, 2018
Google Cloud data engineer,
Google on Coursera 2017
Neural Networks and Deep
Learning, deeplearning.ai 2017
Machine Learning, Stanford
University on Coursera, 2017
Managing an agile team,
University of Virginia on Coursera,
2017
Cloudera Certified Associate,
Spark and Hadoop developer,
2017
Statistical Inference, Johns
Hopkins University on Coursera,
February 2017

MAIN PERSONAL PROJECTS
Miralytics. Free and available on www.miralytics.social
Forex Trading service based on machine learning. Private.
Miralytics, is an analytical tool able to measure the Italian people sentiment about the main politicians
as well as any kind of news posted by the main italian newspaper. The tool continuously monitors and
analyses the social media pages of more than 10 politicians and 6 newspapers, classifying any post
and comment. At the moment more than 100000 posts have been analysed.
I have implemented all the parts of the project, from the: data engineering part, to get and store posts
and comments to the machine learning part to classify posts and comments, I overstretched myself to
also build the user interface (Angular), as well as the backend-api .The tool is hosted on Google Cloud.