Tony Downward
Market Modelling & Decision Science Lead at Contact Energy Ltd
Auckland, Auckland, New Zealand
Summary
👤
Senior🎓
Top SchoolTony Downward is a Market Modelling and Decision Science lead with eight years of industry experience and over a decade in academia, specialising in operations research for electricity markets and multi-horizon decision-making. He combines deep expertise in optimization, machine learning, forecasting and geospatial/network modelling with hands-on software development—most notably as lead developer of the Julia packages JADE and JuDGE that operationalise stochastic hydro scheduling and multi-horizon investment under deep uncertainty. His work has influenced New Zealand market policy and commercial clients including the Electricity Authority, Transpower and major generators, and has been applied to 100% renewable system strategies and pumped-storage assessments. An experienced supervisor and communicator, Tony translates complex models into actionable insights and visualisations for both academic and commercial audiences. Perhaps less obvious: he blends rigorous theoretical publication record with practical tool-building, bridging cutting-edge research and decision-ready software used in national market analysis.
8 years of coding experience
Auckland Grammar SchoolDoctor of Philosophy - PhD, Engineering Science, Doctor of Philosophy - PhD, Engineering Science at The University of Auckland
Github Skills (34)
decomposition10
stochastic10
markov-decision-processes10
stochastic-processes10
julia10
optimization10
capacity9
factorization9
dynamic-programming9
stochastic-differential-equations8
integer8
expansion8
pogo8
mixed-integer-programming7
mathematical6
Programming languages (1)
Github contributions (5)
reganbaucke/JuDGE.jl
Oct 2018 - May 2022
An interface for solving a stochastic capacity expansion problem via a Dantzig-Wolfe decomposition algorithm
Contributions:177 commits, 10 PRs, 156 pushes in 3 years 7 months
stochastic-differential-equationsdecompositionsolvingfactorizationdantzig-wolfe-decomposition
adow031/PoGO.jl
Jun 2022 - Jan 2023
PoGO is a Julia package that facilitates automatic piecewise linearization of various non-linear functions within JuMP models.
Contributions:1 review, 50 commits, 2 PRs in 7 months
julia-packagejumpoptimizationnon-linearlinear
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.