Data Scientist/Researcher/Software Developer

1/2016 – 2/2016

Entwicklung der Bitcoin zahlungssysteme.
Implementation of bitcoin payment system for microworkers.

JavaScript-Frameworks, JavaScript

9/2015 – 12/2015

Full stack software development with Javascript (React JS) and Python, Postgres, AWS.

Postgresql, Python, JavaScript

2/2015 – 7/2015

Software consultancy, advice on a decentralized automated remote micro work platform. Javascript code review, detecting bugs, re-factorization, and technology, solution advice. Specification of a blockchain and smart contracts using Thelonius blockchain framework, decerver distributed application server. Reputation system for workers based on Bayesian update. http://microwork.io/

JavaScript-Frameworks

2/2015 – 7/2015

Implementing backend functionalities using php and mysql and performing data analysis using Python, data reporting, exporting using Propel ORM PHP. Analyzing the effect of delivery time on conversion, conversion of views to cart add, warehouse performance analysis, delivery service performance analysis, brand statistics. Implemented a generic data analysis framework in Python where one can stream data using Amazon Kinesis and connect to different databases and load and get data from them in a black box manner, perform data analysis, use machine learning techniques with visualisation. Wrote a generic web scraping tool and applied it to collecting shipping data from the shipping companies using Python. Python, PHP, MySQL, Amazon Web Services. http://www.sport-thieme.de/

Mysql, Python, Cloud Computing, PHP

8/2014 – 1/2015

Design and implementation of a crowd-founding platform where the decisions are to be made based on collective intelligence. Managed a junior developer. Basic web application functionalities (login, register, etc.), 3 types of voting methods, database methods/ functionalities, user profiles are designed and implemented together with new phenomena such as amplification/decay mechanism for propagation of a decision/suggestion, statistical significance for an election to finalise (multinomial Wilson score, covering collections, Dirichlet distributions - multivariate beta), a greedy algorithm for faster voting under user experience constraints, expectation maximization approach to gain skill points from decision making and ensure accurate decision making, ranking methods, automatic creation of elections. A model of reputation for each individual is designed where comment rankings(Wilson score), consensus on decisions with high skill point users, task completion rate, social network centrality measures in supply-chain context is used. Explored applications of entropy and random fields to life cycle analysis. Tech stack: NodeJS, Python, neo4j, Zero RPC. Testing with mocha and supertest.

Python, JavaScript-Frameworks, JavaScript

8/2014 – 1/2015

Design and implementation of a crowd-founding platform where the decisions are to be made based on collective intelligence. Managed a junior developer. Basic web application functionalities (login, register, etc.), 3 types of voting methods, database methods/ functionalities, user profiles are designed and implemented together with new phenomena such as amplification/decay mechanism for propagation of a decision/suggestion, statistical significance for an election to finalise (multinomial Wilson score, covering collections, Dirichlet distributions - multivariate beta), a greedy algorithm for faster voting under user experience constraints, expectation maximization approach to gain skill points from decision making and ensure accurate decision making, ranking methods, automatic creation of elections. A model of reputation for each individual is designed where comment rankings(Wilson score), consensus on decisions with high skill point users, task completion rate, social network centrality measures in supply-chain context is used. Explored applications of entropy and random fields to life cycle analysis. Tech stack: NodeJS, Python, neo4j, Zero RPC. Testing with mocha and supertest.

Python, JavaScript-Frameworks, JavaScript

7/2014 – offen

Analysis of prostate cancer data set using R.
Statistical tests for comparing two biopsy methods, t-test, Mc Nemar test, Kolmogorov-Smirnov test, Wilcoxon ranked sum test, Barnard, chi-square, Fisher exact, binomial test.

5/2014 – offen

Development of an ETF trading engine to trade electronic currencies using Mastercoin, Ethereum and Bitshares. Python development using Bitcoin, Mastercoin APIs. Ethereum scripting.

4/2014 – 7/2014

Web analytics and conversion rate optimization, Bayesian A/B testing, writing Python and R scripts for Bayesian update and testing of campaigns, using the conversion funnel of the company, Ad/Banner impressions, Landing Page and Offer Impressions.

11/2013 – 1/2014

Email processing module for IMAP accounts: Implementation of regular expressions, a trie for keyword extraction in JavaScript client code.

3/2012 – 10/2013

Contributed to the development of a semantic search engine for retrieval of mathematical expressions. I implemented Java packages for concurrent usage of XML and RDF databases, a Perl module for parsing mixed XQuery and SPARQL queries for querying RDF and XML databases concurrently. Application of natural language processing techniques to mathematics articles, terminological unit identification, named entity recognition: Probabilistic graphical models (HMM, CRF, MEMM), clustering. Creating entries for a semantic mathematics glossary for populating RDF databases with OMDOC Ontology. Four weeks talk; Introduction to Algebraic Geometry and Resolution of Singularities given at the seminar course “Singular Learning Theory” organized by Machine Learning Group, TU Berlin.

10/2010 – 2/2012

I worked as a researcher in symbolic computation group led by Josef Schicho, inside the project Radical Parameterization of Algebraic Curves. I implemented numerous algebraic geometry and computer algebra algorithms using Sage, Python. I made research on the improvement of various number theoretic and algebraic geometric algorithms. I wrote several technical reports. I have given talks and attended several conferences.

3/2010 – 8/2010

I worked in 2 software projects for developing software to manage large number of printers via a user interface. Client-server programming.
I completed a C# package for performing disk operations in printers, using PJL queries, socket programming, developed Java classes for monitoring jobs sent to printers, wrote a Python script for analyzing the performance of the printer while the software is running, producing charts using Python packages Scipy/Matplotlib, wrote a Java library for achieving pdf encryption.

3/2009 – 1/2010

I performed statistical programming using Python and R, statistical analysis of genomic data, case-control studies. I completed data quality control, principal components analysis of DNA data collected from individuals belonging to Spanish population in order to detect susceptibility variants of various diseases by using Python and R, made research on gene-gene interaction, implemented existing statistical and machine learning techniques in the literature, designed, implemented and tested own statistical hypotheses test, obtained positive results. (p values less than 0.05 for 47 out of 50 epistasis models.)

7/2008 – 9/2008

I took part in the design and analysis of a SHA-3 candidate hash functions. I did various experiments using Maple and performed differential, linear cryptanalysis, man-in-the middle attack, checked pre-image resistance.

9/2007 – 3/2008

Collaborated in the generation of the patent: Method and System to Provide Fine Granular Integrity to Digital Data.
I worked as a researcher on the design of a system to detect unauthorized modification of audit logs and data. Solving the problems posed by the software development team concerning security by using cryptographic primitives and mathematics. As well as reading articles and books in cryptology and information security, I could design my own method. Collaborated in the defense of the patent; “PCTES0300439-EP20030818388: System for Implementing a Game of Chance over a Communications Network”

6/2007 – 9/2007

I worked on providing data privacy in private information retrieval systems. I read articles on information security and cryptology, I wrote a draft on PIR.