Nids Homepage

Hi my name is Francesco Nidito and I got my Ph.D. in Computer Science from the Dipartimento di Informatica of Pisa University in 2008.

During my Ph.D., I worked on topology control and data load-balancing in wireless sensor networks. My Ph.D. thesis is about the influence of non-uniformity in wireless sensor networks: if you are not able to sleep, this is the link.

While still a student, I started working for R&D office in Pisa where I worked on multimedia indexing (mainly images) and news search.

For the Image Search, my job was to create and implement new ideas/algorithms for the processing of multimedia content and the features extraction from such documents.

For the News Search, I was one of the maintainers of part of the crawling infrastructure for the videos used in News search.

Moreover, I implemented various systems to monitor the execution of our processing infrastructure: 24/7 thousands of machines.

In November 2009, I joined the Search Technology Center of London working for Bing… yes, another search engine.

Just to maintain the good habits, I continued working on Multimedia search, focusing on the relevance/ranking of the results and on the triggering logic of the Multimedia instant answer (the one that presents images and videos inside the page of web results).

In 2010, I become the Development Lead of the Multimedia relevance team in London.

At the end of 2014, I resigned from Microsoft and concluded my adventure in the Search Engines land…

… not really as my new job is in Bloomberg LP with focus on News Search relevance 😉

After more than 3 years in Bloomberg LP and almost 10 in the United Kingdom, I moved back to Italy and started working with some old friends of mine at Viralize.

Aside from my job and past Ph.D., I am also interested in Operating Systems topics such as multi-threading and concurrency; scripting and meta-programming (both template meta-programming and generative programming).

You can find me on LinkedIn, and Reddit.


I like to write extendible code which needs as little changes as possible to add features. That is totally doable in Perl enabling the loading of modules at run time. To know how to do that, you can read Dynamically loading modules in Perl at run time.

Some time ago I had a discussion about tuples and pairs in C++ and while the discussion didn't reach a conclusion, that made me thing and write If you cannot name it, it shouldn't exist!

After reading some critiques on the TIOBE and PYPL indexes for programming languages popolarity, I decided to experiment with Liked In data to have a bit of fun. My results can be found in The Linked In Languages (Popularity?) Index.

Still in the spotlight my first post of 2016 about the worst Perversions, Cults and Fetishes of the Information Age. It is supposed to be fun and, if you recognize yourself in it, don't be too upset and do have a laugh 😊