Case Study

Example 1

Client

 A leading PC security software company

scope

Functional and technical design, development, quality assurance, documentation, on going development and support

period

 2003 to date – on-going

type of software

 Core agent software, behavioural analysis of virus/Trojan/malware behaviour, malware behavioural blocking, operating system call level development

Skills

 Windows, UNIX and LINUX kernel skills, algorithm

Notes

 The client is the world’s largest supplier of consumer IPS software.
They required a policy driven, behavioural (i.e. non-signature based product), that could both block a high proportion of software attacks, whilst not being “noisy” or intrusive for home PC users.
The project involved complex operating system design and development skills, in conjunction with a deep behavioural analysis of malware techniques. An exceptionally high level of quality assurance was required to support a large number of home PC downloads.
The client now has approximately 1,000,000 users of the product.

Example 2

Client

 A major consumer loan company

scope

Design, development, implementation of real time Risk analysis, loan approval system platform and straight though processing, including design, technical specification, development, qa, documentation, on going development and support

period

 2007 to date – on-going

type of software

 Consumer loan and risk management system

Skills

 .NET, C#, SQL Server, Dundas OLAP, BI Datawarehousing, AI/ML, statsistical analysis

Notes

 The client is major, highly successful consumer loan organization providing web-based loans. Their products are functionally leading edge, delivering real time, web-based loan approval. This has required innovative and highly creative development of a complex combination of risk techniques including:
credit risk application scoring, behavioral scoring, fraud scoring, collection scoring, credit limit modelling, yield modelling, LGD modelling, statistical data mining, portfolio analysis supervision, business strategy, decision development.
Algorithms used include – Logistic regression, Neural Networks, Support vector machines, Naive Bayes, Nearest Neighbor, Decision Trees, Cluster analysis, Survival analysis (Cox regression, Caplan-Meier), Discriminant analysis, Factor analysis, Principal Components analysis, Partial Least Squares regression, Markov chains, Monte Carlo, Boosting algorithms.

Example 3

Client

 A supplier of credit card fraud prevention software, and complex data quality management to the finance and banking sector

scope

Function design, technical specification development, QA, documentation, on going development and support

period

 2004 to date – on-going

type of software

 Data quality analysis tool for identifying underlying movements, structural changes in fast moving data, (bank trading/ credit card etc), utilising entropy analysis, mutual association technique, dynamic relationship trees, vectorization and complex/real time OLAP functionality

Skills

 Prolog, java, C++, complex behavioural algorithms

Notes

 The client is a leading edge, mathematical software house specialising predictive analytics and the application of Inductive Logic programming to complex financial sector applications

Other development projects in which the Zoral team has considerable experience include:

  • development of complex, low-latency, large-scale Data Warehousing systems utilizing Ab Initio (ETL engine) and C++
  • design and development of meta data repository and executable meta data driven rules engine
  • development of high speed, large volume data analysis and OLAP
  • developing metadata/rules driven, multi-tier, Java based GUI and visualization engines and components
  • development of custom parsers, interpreters
  • development of automated code conversion engines to convert from a source language to a target language with minimal manual intervention
  • ODBC/JDBC drivers and interfaces in C++ for vector RDBMS
  • Simba/MDX drivers and interfaces for OLAP in C++ and Java
  • development of a comprehensive security environment and host intrusion prevention system, including custom security drivers, applications, methods and procedures and implementation for a Windows 2000 operating system based topology. Implementation operating system kernel level security, databases and applications security Software utilized: C++, Assembly, Java, various RDBMS engines
  • development and implementation of real-time trading systems for a major investment bank utilizing C++/Unix
  • designed and developed a secured system for Internet payments and transfers (BSD Secure shell), for the middle office of an Investment Bank