Career Profile

I’m a seasoned machine learning professional, with a proven leadership record and over 10 years experience. My expertise lie in harnessing the power of data to drive business transformation - pushing the limits of what's possible on the technical side to carve out efficiencies on the business side.

I'm currently Vice President of Data Technology at Compass Digital Labs, where I grew a data practice from the ground up (consisting of Data Science, Business Intelligence, Advanced Analytics, Data Engineering, and Machine Learning Lab functional groups). Equal parts formulation, innovation, and implementation, my teams interface with Fortune 100 clients to boost their retail food operations with data-supported and data-driven strategies. We are well on our way to achieving 9-figure top line lift from our primary data projects, supporting over $20B of business from our parent company Compass Group North America. Previously, i worked at Crescendo Technology (subsidiary of Pinnacle Sports), where I was the first Data Scientist on staff. I was tasked with developing real time breaking news algorithms, as well as real-time NBA gambling machine learning models.

My PhD and PostDoc work focused on modelling the impacts of poly unsaturated fatty acids on trophic transfer efficiencies in fresh water planktonic food webs. All this through fundamental mechanistic models, culminating in a cybernetics approach. During this time, I consulted with the governments of Canada, USA, South Africa, and China on environmental and ecological model building.

Experiences

Vice President, Data Technology

October 2020 - Present
Compass Digital Labs, Toronto
Strategy & Partnerships

Restructured team and shifted focus to navigate the Covid-19 global pandemic. Partnership with sibling teams and other analytically-minded groups across the company is a key driver in our current data pursuits. There is no economic precident for current fiscal dynamics. Using data for driving forward, as a feedback loop, a means of automation, or simply a tool to play devil's advocate, the apptetite for data-driven decisions is at an all-time high (fall, 2020).

Senior Director, Data Intelligence

July 2018 - September 2020
Compass Digital Labs, Toronto
Built Data Products and Platforms.

Shifted away from building custom one-off models/reports/analyses/pipelines, and towards self-serve products and platforms. These products continue to drive both revenue and cost-savings. During this period, spun up Research and Development initiatives, encouraging employees to step outside the confines of their well defined projects for the purposes of learning/experimenting/enhancing.

Director, Data Intelligence

November 2017 - July 2018
Compass Digital Labs, Toronto
Drove full-stack Data Practice.

Inherited (and dismantled) Business Intelligence department, forming Data Intelligence group (Data Science + Business Intelligence). Revamped Business Intelligence practices, splitting focus between reporting and Data Engineering (ETL, pipelineing, data reliability etc.). Stood up several (successful) full-stack proofs of concept for both internal and external use, demonstrating Art of the Possible style thinking.

Senior Data Scientist

November 2016 - November 2017
Compass Digital Labs, Toronto
Built out Data Science team and Data Science practice.

Built Data Science team to work on data cleansing and State of the Art algorithms to drive value and data-driven mindsets within the organization. Worked closely with Marketing, Field Services, and Operations to deliver one-off data-driven narratives using modeled data built off cleansed data.

Data Scientist

July 2015 - November 2016
Crescendo Technology (Pinnacle Sports), Toronto
Built machine-learning algorithms for real-time consumption.

Built State of the Art news listening algorithms for contextually relevant news, while effectively filtering out noise, false signals, and amplification. Revamped existing probabilistic real-time NBA gambling models with a machine learning approach.

Post-Doctoral Research Scientist

September 2012 - July 2015
University of Toronto, Toronto
Cybernetics & Academic Consulting

Built on graduate work, expanding mathematical models with intra-individual complexity, and extensibility to be plugged into agent-based environments. During this period, I also participated in extensive consulting work, with subject matters spanning aquatic oil spills (USA), holistic ecological modelling (South Africa), pre-construction knowledge sharing (China).

Projects

Side projects are a fantastic way to learn a new technology, sharpen your existing skills, and learn a new domain. A select number of my projects are presented below.

GSS - An on premises information display system custom tailored for Scarborough Gurdwara. Includes cloud hosted admin panel, responsive HTML5 app, and information kiosks. Status: Live. Established 2007.
BallQuant - A.I.-Advisor for sports gambling. Accuracy over 85%, and demonstrated over 1,000% ROI. Status: Open for registration (free trial available). Established 2018.
Pulse - A natural language parsing news listener that indexes current events by subject time, matter, political spin, and news-cycle longevity. Status: Prototype. Established 2019.
Continuum - A temporal and geospatial model, accounting for changes in cost of living, crime, traffic, and various other socio-economic indicators. Status: Prototype. Established: 2019.
Evolver - A.I. stock picking advisor built on technical trading foundations, trained using a genetic algorithm. Returns over a 12 year period outshine the S&P500. Status: MVP. Established 2020.

Skills & Proficiency

Problem Solving

Leadership

Mathematical Modelling

Machine Learning

Python

R

Team Building

Business Strategy

Publications

An integrative methodological framework for setting environmental criteria: Evaluation of stakeholder perceptions.

2017
Ecological Informatics 48: 147-157.
Kim D-K., M. Ramin, V. Cheng, S. Kaluskar, D. Kobiliris, A. Neumann, F. Ni, T. Peller, G. Perhar, Y. Shimoda, A. Visha, C. Wellen, S. Mugalingam, G.B. Arhonditsis.

Integrative Analysis of Lake Simcoe Watershed (Ontario, Canada) as a Socio-Ecological System.

2017
Journal of Environmental Management 188: 308-321.
Neumann (Gudimov), A., D.K. Kim, G. Perhar, G.B. Arhonditsis.

Using Daphnia physiology to drive food web dynamics: A theoretical revisit of Lotka-Volterra models.

2016
Ecological Informatics 35: 29-42.
Perhar G., N.E. Kelly, F.J. Ni, M.J. Simpson, A.J. Simpson, G.B. Arhonditsis.

Towards the development of an early warning system for ecosystem health: A metabolite-driven Daphnia ecophysiological model.

2015
Ecological Informatics 30: 82-96.
Perhar G., G.B. Arhonditsis.

What has been accomplished twenty years after the Oreskes et al. (1994) critique? Current state and future perspectives of environmental modeling in the Great Lakes.

2014
Journal of Great Lakes Research 40: Supplement 3 1-7.
Arhonditsis, G.B., C.A. Stow, Y.R. Rao, G. Perhar.

Aquatic ecosystem dynamics following petroleum hydrocarbon perturbations: A review of the current state-of-knowledge.

2014
Journal of Great Lakes Research 40: Supplement 3 56-72.
Perhar, G., G.B. Arhonditsis.

Ecological consequences of largemouth bass (Micropterus salmoides) and bluegill (Lepomis macrochirus) in Lake Kawahara-oike, Nagasaki, Japan.

2013
Ecological Informatics 18: 149-161.
Hossain, M., G. Perhar, G.B. Arhonditsis, T. Matsuishi, A. Goto, M. Azuma.

Modeling zooplankton growth in Lake Washington: A mechanistic approach to physiology in an eutrophication model.

2013
Ecological Modelling 258: 101-121.
Perhar, G., Arhonditsis, G.B., Brett, M.T.

Modelling the role of highly unsaturated fatty acids in planktonic food web processes: Sensitivity analysis and examination of contemporary hypotheses.

2013
Ecological Informatics 13: 77-98.
Perhar, G., Arhonditsis, G.B., Brett, M.T.

Modelling the role of highly unsaturated fatty acids in planktonic food web processes: A mechanistic approach.

2012
Environmental Reviews 20: 155-172.
Perhar, G, G.B. Arhonditsis, and M.T. Brett.

Examination of the role of detritus food quality, phytoplankton intracellular capacity, and zooplankton stoichiometry on planktonic dynamics.

2012
Ecological Informatics 11: 76-89.
Perhar, G., and G.B. Arhonditsis.

Examination of the effects of nutrient regeneration mechanisms on plankton dynamics using aquatic biogeochemical modeling.

2012
Ecological Modelling 240: 139-155.
Ramin, M., G. Perhar, Y. Shimoda, and G.B. Arhonditsis.

Our current understanding of lake ecosystem response to climate change: what have we learnt from the north temperate lakes?

2011
Journal of Great Lakes Research 37: 173-193.
Shimoda Y., E. Azim, G. Perhar, M. Ramin, M. Kenney, S. Sadraddini, A. Gudimov, and G.B. Arhonditsis.

The effects of seston food quality on planktonic food web patterns.

2009
Ecological Modelling 220: 979-997.
Perhar, G, and G.B. Arhonditsis.

Bayesian calibration of mechanistic aquatic biogeochemical models and benefits for environmental management.

2008
Journal of Marine Systems 73: 8-30.
Arhonditsis, G.B., D. Papantou, W. Zhang, G. Perhar, E. Massos, and M. Shi.

Addressing equifinality and uncertainty analysis in eutrophication modelling.

2008
Water Resources Research 44: W01420.
Arhonditsis, G.B., G. Perhar, W. Zhang, E. Massos, M. Shi, and A. Das.