I work as a scientific software engineer at my own company datiaperti based in Rome, Italy. Before taking the leap from being an employee to being self-employed, I worked as a software engineer in many public and private organizations in Italy, Switzerland and Germany. In Germany I worked at Fraunhofer IAIS in EU and German funded research projects and industry projects related to Semantic Web, Machine Learning and Big Data technologies. In Switzerland, I worked as a research associate and software engineer in two FP7 research projects about data integration using Semantic Web technologies at the E-Government Institute of the Bern University of Applied Sciences. Before moving abroad I have worked in Italy as a consultant in software development projects at various multinational companies such as Ericsson, Marconi Communications, Accenture, and Italian government institutions. I hold an M.Sc. in Physics from the Sapienza University of Rome with a specialization in Atmospheric Physics and Remote Sensing. I have (co)authored papers on Semantic Web, E-Government and Remote Sensing. My main interests are computational physics, artificial intelligence, algorithms, computer vision, open source software development and open data.

Work experience

Fraunhofer IAIS (2015-12-01 – 2019-11-30), Sankt Augustin (Germany)

I have been working as project leader and software architect in EU and German research projects and industry projects based on Big Data, Semantic Technology and Machine Learning. As a software architect I have designed the architecture of the Smart, Green and Integrated Transport pilot for the Big Data Europe H2020 project using open source frameworks such as Apache Kafka, Apache Flink, and Elasticsearch. I have also acquired experience in using Docker containers for deployment. I have led an industry project for a Spanish customer to implement and deploy the back-end of their online e-commerce platform and additional services based on Apache Solr and Apache Cassandra and machine learning algorithms for recommendations and price prediction. As a project leader I have applied agile methodologies in both research and industry projects for project planning and management, cost estimation, development and continuous integration using the full set of tools provided by Atlassian.

E-Government Institute of the Bern University of Applied Sciences (2012-09-01 - 2015-10-31), Bern (Switzerland)

I have been involved in two European FP7 projects both led by the Bern University of Applied Sciences. In the first project, Fusepool, I developed Java OSGi components for the transformation of XML or tabular data into RDF and for the disambiguation and interlinking of entities. The second project, Fusepool P3 was aimed at providing components to implement workflows for Linked Data Platforms. I participated in the writing of the proposal and in the development of the software. Each platform components could communicate with others through a REST interface. My work has been focused on interlinking, based on the Silk framework and in the use of geographical data. I developed also a semantic web application to visualize geographical data from a Linked Data Platform and search points of interest in a map. In both projects I participated in the writing of deliverables and published papers as author or co-author in peer-reviewed journals.


  • Josep Maria Salanova Grau, Evangelos Mitsakis, Panagiotis Tzenos, Iraklis Stamos, Luigi Selmi, Georgia Aifadopoulou: Multisource Data Framework for Road Traffic State Estimation, Journal of Advanced Transportation, Hindawi
  • Ivan Ermilov, Axel-Cyrille Ngonga Ngomo, Aad Versteden, Hajira Jabeen, Gezim Sejdiu, Giorgos Argyriou, Luigi Selmi, Jürgen Jakobitsch, Jens Lehmann: Managing Lifecycle of Big Data Applications, International Conference on Knowledge Engineering and the Semantic Web, Springer
  • L. Selmi, A. Neuroni, Fusepool P3: A Linked Data Platform for Open Government Data, IOS Press
  • M. Kaschesky, L. Selmi: Fusepool R5 linked data framework: concepts, methodologies, and tools for linked data, Proceedings of the 14th Annual International Conference on Digital Government Research
  • P. Castracane, L. Selmi, S. Casadio, M. Cacciani and G. Fiocco, Ground-based remote sensing of wind, temperature and aerosol backscattering in an urban environment during different atmosphericstability conditions, Ed. Physics and chemistry of the earth. Part B: Hydrology, oceans and atmosphere, Elsevier Science.

Presentations and awards

  • 23-27 May 2022 - Living Planet Symposium 2022, Bonn (Germany) - Land Use and Land Cover Classification using CNN Deep Learning Architectures (Poster Session)
  • 25-27 January 2022 - The 4th International Electronic Conference on Remote Sensing - Flood mapping using Sentinel-1 SAR imagery and the ESA SNAP S1 Toolbox (Best Presentation Award)

Courses and Certifications

  • 2021-09-14 - Echoes in Space - Introduction to Radar Remote Sensing, EO College
  • 16-20 November 2020 - 2nd Joint Training Course on Atmospheric Composition, EUMETSAT, ESA, ECMWF
  • 2020-08-10 Algorithms, Part 1. First part of the course taught by Kevin Wayne and Robert Sedgewick and offered by Princeton University on Coursera. This part of the course covers elementary data structures, sorting, and searching algorithms in Java.
  • 2020-03-09 Algorithms, Part 2. Second part of the course taught by Kevin Wayne and Robert Sedgewick and offered by Princeton University on Coursera. This part of the course focuses on graph and string-processing algorithms in Java.
  • 2019-07-12 - Cryptography I. An online course about symmetric and public-key cryptography taught by Prof. Dan Boneh and offered by Stanford University on the Coursera platform. I achieved a grade of 100 %.
  • 2018-07-22 - Deep Learning Specialization. An online specialization on Deep Learning offered by deeplearning.ai on the Coursera platform and taught by Prof. Andrew Ng. It is made up of five courses: 1) Neural Networks and Deep Learning, 2) Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization, 3) Structuring Machine Learning Projects, 4) Convolutional Neural Networks, 5) Sequence Models. I achieved a grade of 100 % in all the five courses.
  • 2016-10-30 - Machine Learning. An online course offered by Stanford University on the Coursera platform and taught by Prof. Andrew Ng. The course provides a broad introduction to machine learning, datamining, and statistical pattern recognition. Topics include: (i) Supervised learning (parametric/non-parametric algorithms, support vector machines, kernels, neural networks). (ii) Unsupervised learning (clustering, dimensionality reduction, recommender systems, deep learning). (iii) Best practices in machine learning (bias/variance theory; innovation process in machine learning and AI). I achieved a grade of 98.8 %.
  • 2014-12-22 Introduction to Logic. An online course offered by Stanford University and taught by Prof. M. Genesereth. The course is a rigorous introduction to Logic. It covers the encoding of information as logical sentences, logical reasoning,and an introduction to logic technology and its applications. Logics covered include Propositional Logic and Relational Logic. I completed the course with distinction.


Italian is my mother tongue. I speak English fluently as it’s the language I use the most since 2010. I can understand quite a bit of German, French and Portuguese. I got a Cambridge First Certificate English (Level B2 of the EU Reference for Languages) in 2010 and a telc Zertifikat Deutsch B1 in 2017.

Other skills

I like being in the nature, most often in the Apennine Mountains in Central Italy, not far from Rome. I have a 1° degree FIPSAS/CMAS diver license and I have been practicing mountain climbing and skiing.