• Our Team

    The team at  Airmed use Machine Learning, ‘Big Data’ toolsets and statistical modelling to unleash the power of predictive analytics. We provide a blend of statistical, data and computer science expertise to enable data collection, processing, storage and visualisation, data mining, forecasting and improved decision-making – essentially “turning data into gold”.

    Our background involves health research and handling broad and large datasets and using a wide range of IT solutions, including business intelligence, statistics, SharePoint, SQL, GIS, Internet of Things (IOT) and bioinformatics with a focus on IOT, machine learning and predictive analytics.


    Benoit Auvray


    Benoit Auvray is a director and the lead data scientist at Airmed, and also has a part time position as a senior research fellow in the department of Mathematics and Statistics of the University of Otago. Benoit has experience in data science and predictive analytics using big data applied to a wide variety of areas, including quantitative genetics and genomics (his particular area of expertise), agricultural applications, traceability and food safety, survey analysis, tourism, education and others.

    Originally from Belgium, Benoit obtained his MSc from Gembloux University in quantitative genetics in 1999. During his Master he helped develop software at the University of Georgia (USA) used for genetic evaluation of bivariate threshold-linear traits. After graduating, Benoit worked in agricultural mechanisation and statistical analysis of clinical trial data before coming back to quantitative genetics and helping develop the current dairy cattle evaluation system of the Walloon region in Belgium. In 2003, he moved to New Zealand to work at AgResearch Ltd on projects to map QTL in sheep and integrate genomic information into genetic evaluation systems for Ovita Ltd. Since then he has helped to create a new 2-stage genomic evaluation system for the dual purpose NZ sheep industry using genotypes from low density SNP arrays, being mostly responsible for the theoretical development and first practical implementation of the current system (‘SHEEP5K’, commercialised by Zoetis, Inc.). In 2014 he moved to the Mathematics and Statistics department at the University of Otago where he is now leading a project aiming to develop a new genomic evaluation system for NZ sheep, through Beef+Lamb Genetics Ltd, integrating in a single-step evaluation animal performance, pedigree and genomic information to estimate the breeding potential of animals. The new system is expected to be in place for commercial use in 2017 and will replace the current Sheep Improvement Limited national evaluation system for dual purpose sheep.

    During his career and through creating Iris Data Science in April 2013 in partnership with Greg Peyroux and working in a more commercially oriented environment, Benoit has developed a strong business understanding, customer focus and commitment to produce excellent commercially sensible solutions to his clients to complement his 16 years of experience in research.

    In his spare time, Benoit enjoys looking after his 4 children, practicing and teaching combat sports (Benoit is an active brown belt competitor in Brazilian Jiu Jitsu), travelling, playing board games, cooking, reading, writing statistical models for sport predictions, juggling and the occasional hike.

    Alesha Smith


    Dr Alesha Smith BSc, MSc (Distinction) (Otago), PhD (Queensland) 


    Alesha has more than 10 years’ experience in the field of health as an academic and consultant. She has provided research solutions and outputs for a wide range of organisations including national and global health bodies, government departments, universities and private companies.

    Greg Peyroux


    Mr Greg Peyroux is Director and Airmed and has over 15 years’ experience as an ICT leader, manager and strategist. He has managed, designed and implemented a rich set of tools and systems across the spectrum of IT solutions, including financial and employee systems projects, decision support systems for large pastoral sector stakeholders, through to significant national research high performance computing initiatives.

    Greg brings experience in Information Security, including as a security manager for the Crown Research Institute, AgResearch and being a member of the New Zealand ISO Subcommittee 27 – Information Security. He has developed good capabilities in data management, having prepared the Data Management Strategy for the New Zealand Greenhouse Gases Centre in Palmerston North and managing the design and construction of its data repository.

    Greg is also an experienced ICT manager and people leader, including significant periods of time as an acting CIO at AgResearch, as a Software Development Manager and as an Information Solutions Manager encompassing responsibilities for Bioinformatics, High Performance Computing (HPC), Geographic Information System (GIS), SharePoint, Nintex, SQL Server and BizTalk.

    Outside of work Greg is involved in various community activities: as the Musical Director of a University choir, managing a team at a local Junior Football club, as treasurer for a national registered charity and is an accomplished organist. As a choral organist, recitalist and conductor, Greg has performed in many distinguished locations, including Sydney, Melbourne, Ely, London’s St Paul’s Cathedral, Dublin, St Andrews, Vienna and Leon.

    Masha Mikhisor

    Computer Vision and Machine Learning Engineer

    Masha Mikhisor is a computer vision and machine learning engineer. Masha has experience in applying different supervised and unsupervised learning methods for real time image recognition tasks.

    Masha’s expertise in image processing libraries such as OpenCV is advantageous for images prepossessing and extracting features for machine learning algorithms. Her technical skills include code optimizing for parallel processing which is very important for performing complex processing tasks on large sets of images in a reasonable amount of time.

    Masha obtained her MS degree in Moscow Institute of Physics and Technology in applied physics and mathematics in 2010. As her Master’s project, she developed an interactive virtual reality scene for a downhill skiing training system. In 2013 she moved to New Zealand to obtain her PhD in the computer science department of Otago University. During her PhD studies, she developed a real time 3D face tracking system based on two stereo cameras. In this system a semisupervised learning method is used to learn the appearance of a tracked person. The learned appearance model is then used to distinguish this person from other people present in the scene and recover the tracking promptly if the tracked person leaves and then reappears in the scene.

    In her spare time, Masha enjoys spending time outdoors with her family, hiking, biking, cross-country running, orienteering, snowboarding. She enjoys multi-day hikes and endurance running events. She was one of the organizers and event directors of Dunedin parkrun, weekly running event in Dunedin Botanic Garden.

    Dr Manjula Devananda

    Health Data Scientist

    Manjula Devananda is a Health Data Scientist and enjoys working with health data and programming. Her expertise in Health informatics and data science is important to address optimising system that aid decision making in health care domain. Manjula’s research experience include working on a six-month project at University of Saskatchewan in 2009, R&D project engineer at Wipro Technologies for more than two years and a Ph.D. in Information Science from University of Otago conferred in 2019.

    During her Ph.D., Manjula attempted to predict the upcoming workload from long-term conditions at a primary health centre. This work also learned, using a Bayesian inference model, that the knowledge of patients about their medications running out influence their decision to  visit for a follow-up. She also explores the changes required at a practice due to new policies and procedures at a national level.

    In her spare time, she enjoying dancing Bharatanatyam, a South Indian classical dance, and Carnatic music.