Florin Chelaru

Florin Chelaru

CTO @Twinfog

  I'm a software engineer with a Ph.D. in Computer Science from the University of Maryland and industry experience at Microsoft, Facebook and MIT. My expertise ranges from designing complex machine learning software for big data analysis, to high end UX-driven tools for data visialization, social platform spam detection and mobile apps.

At present, I am the CTO and primary Mobile Development engineer at Twinfog.


florin [at] twinfog [dot] com


Getting ready to launch Twinfog, an augmented reality social app for local communities.

Ask for help or opinions from people in your area, and set a deadline for answers – to make sure they are relevant.


University of Maryland, College Park

PhD, Computer Science2011 - 2015

Advisor: Héctor Corrada Bravo
Cumulative GPA: 3.89/4.0
Thesis: Epiviz: Integrative Visual Analysis Software for Genomics

University Al. I. Cuza, Iași, Romania

BSc, Computer Science2004 - 2008

Advisor: Liviu Ciortuz
Cumulative GPA: 9.45/10.0
Class rank: 8 of 176
Thesis: Artificial intelligence in Computer Go


Metaviz: Integrative visualization for metagenomics

F. Chelaru*, J. N. Paulson and H. C. Bravo, In preparation.

Epiviz: a view inside the design of an integrated visual analysis software for genomics

F. Chelaru* and H. C. Bravo, BMC Bioinformatics, 2015;16(Suppl 11):S4. doi:10.1186/1471-2105-16-S11-S4.

Epiviz: interactive visual analytics for functional genomics data

F. Chelaru*, L. Smith, N. Goldstein, and H. C. Bravo, Nature Methods, vol 11, no. 9, pp. 938-940, Aug. 2014.

epivizr: R Interface to epiviz web app

H. C. Bravo*, F. Chelaru, L. Smith and N. Goldstein, Bioconductor Package, 1.4.2.

Combining old-fashioned computer go with monte carlo go

F. Chelaru* and L. Ciortuz, Proceedings of the 2008 10th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing, SYNASC 2008, 2008, pp. 216–222.

The general Chinese remainder theorem

S. Iftene* and F. Chelaru*, International Scientific Journal of Computing, vol. 6, issue 1, pp. 44-50, 2007.


Twinfog Inc.

CTOJun 2016 - Present

Designing and implementing the Twinfog cross platform mobile app using Xamarin (C# .NET) for the UI, and ASP.NET MVC, SQL Server and Azure Cloud Services for the back-end.

MIT Computer Science and Artificial Intelligence Laboratory

Postdoctoral AssociateAug 2015 - Jun 2016

Expanded on the work done for my Ph.D. by designing a series of open-source visualization libraries for genetic variants analysis. Base library code available here:

Center for Bioinformatics and Computational Biology

Research AssistantJan 2011 - Jun 2015

Designed and developed software to aid in the analysis and exploration of Big Data, in particular high throughput sequencing – using Machine Learning and Visualization to comprehend relationships and correlations between epigenetic mechanisms and gene regulation.

Rocket Fuel - Artificial Intelligence Team

Graduate Intern2014 - Jun-Sep
2013 - Jun-Sep

Designed and implemented Machine Learning probabilistic models and Visualization infrastructure for Big Data analysis, in particular for Ad Click Prediction.

Facebook - Site Integrity (Spam Detection)

Graduate Intern2012 - May-Aug

Designed and implemented Machine Learning models for the detection of spam users and content.

University of Maryland Department of Computer Science

Graduate Teaching Assistant2011 - 2013

CMSC702 – Computational Systems Biology
CMSC433 – Parallelism and Multithreading in Java
CMSC420 – Data Structures

Microsoft Inc. - Office Team (Lync Server)

Software EngineerJun 2010 - Jan 2011

Designed database optimization software for improving the performance of the Lync Communication Server.

Microsoft Inc. - Bing Team (Domain Relevance)

Software EngineerSep 2008 - Jun 2010

Designed and developed software for measuring the quality of web search results. Specifically, created Machine Learning models for improving the relevance of the content of text snippets, and Visualization software for monitoring their quality.

Code40 Inc. Romania

Undergraduate Internship2007 - Jul-Aug

Designed and implemented components of a web server application for micro-loans: caching, back-end data validation, error handling.

Open-Source Software

Epiviz (

Aug 2014

A web visualization tool used to aid in the analysis and exploration of large functional genomics data. Technologies used: HTML5, JavaScript (JQuery, AngularJS d3.js, WebSockets), PHP, MySQL, R/Bioconductor, Python.

Epivizr Bioconductor Package

Aug 2014

A package that provides WebSocket communication to the Epiviz web app for interactive visualization of genomic data. Objects in R/bioc interactive sessions can be displayed in genome browser tracks or plots to be explored by navigation through genomic regions.
Authors: H. C. Bravo*, F. Chelaru, L. Smith, N. Goldstein


Feb 2007

A .NET app used to create ISO 9660 Joliet CD/DVD images from folders on the local machine. It currently has ~20K downloads per month.
Technologies used: C#, .NET Framework 2.0.

Technical Skills

OO Low-level Languages

Java (J2EE); Microsoft .NET (C# and the CLR); C; C++

OO and Functional High-level Languages

Python; R/Bioconductor; JavaScript (JQuery), PHP


SQL (MySQL, Microsoft SQL Server); Apache Hive (Hadoop)


Romanian (native)



Héctor Corrada Bravo

Héctor Corrada Bravo

Department of Computer Science
Biomolecular Sciences Building
University of Maryland
College Park, MD 20742

hcorrada [at] umiacs [dot] umd [dot] edu


Mihai Pop

Mihai Pop

Department of Computer Science
Biomolecular Sciences Building
University of Maryland
College Park, MD 20742

mpop [at] umiacs [dot] umd [dot] edu


Jack van Ryswyck

Jack van Ryswyck

Artificial Intelligence
Rocket Fuel Inc.
1900 Seaport Blvd
Redwood City, CA 94063

javhar [at] rocketfuelinc [dot] com

Michael Benisch

Michael Benisch

Artificial Intelligence
Rocket Fuel Inc.
1900 Seaport Blvd
Redwood City, CA 94063

mbenisch [at] rocketfuelinc [dot] com