Accomplishments Overview
This page lists projects, publications, presentations, and other achievements the I have been involved with. Feel
free to browse the various social links available (LinkedIn,
GitHub, Google Scholar) to get a rough idea.
Business and Data Science / AI Leadership achievements
Owned the Peak AI Application lifecycle
- Product: Took responsiblity for Peak's standard application offering, and launched Peak's Inventory,
Pricing, and Customer Segmentation product lines. Responsible for all aspects of Peak's suite of standard AI
Applications, including vision, features, and design. Success of these products resulted company-wide Strategy
and
GTM pivot (2021-2024)
- AI Engineering: Responsible for all of the engineering and design teams delivering Peak's AI
Application
offering. Pioneered a modular and extensible framework for application development, prioritising flexibility of
customisation to support rapid iteration while establishing product-market fit (2021-)
- AI Services: Responsible for building and scaling one of the best AI Services teams in Europe. The
team's
responsibilities implementing Peak's applications, as well as providing Insight and Data Engineering services,
and
building bespoke applications, for companies like Nike, PepsiCo, and Ralph Lauren (2019-2024)
- Sales: As part of the senior leadership team, ongoing participation in defining commercial strategy,
including GTM,
pricing, and sales plays. Responsible for delivery of technical pre-sales / solution engineering (2019-2023)
- R&D: Responsible for all AI R&D, including deep/academic R&D; applied/discovery-driven R&D; field
R&D/bespoke development; and ensuring that R&D efforts feed product development and generate value (2017-)
Helped scale a business from ten to hundreds
- Launched, nurtured, and grown business divisions in 3 distinct regions (UK, USA, India)
- Helped grow a business from startup-sized customers and contracts worth thousands, to global enterprise
customers and contracts worth
millions
- Supported on a Series B and a Series C funding round (over $100m in funding raised)
- Built a production line for AI leaders. Developing, training, and promoting high quality data science
managers,
engineering managers, and product managers
- Navigated leadership challenges through periods of both hyper-growth, and industry downturn
Diversity, Equity, Inclusion, and Outreach
- Supported industry DEI efforts, by providing diverse pathways into AI, including internships, fellowships, and
grad schemes
- Championed DE&I causes, striving for effective allyship. Built and retained a highly technical team with
industry leading diversity metrics at all seniority levels (50% gender split in senior roles, 38% gender split
across all technical roles), and maintained a negligible gender pay gap
- Averaged industry leading retention rates despite team of 80+, including a peak YOY retention of over 95%
- Launched external mentoring scheme
Projects and Use Cases
AI Use Cases
I have been responsible for the delivery of AI and Data Science implementations for 80+ businesses. Many are
covered by customer confidentiality, or NDAs, but high level details are shared here.
Sales and Marketing Use Cases
- Predictive Customer Analytics - Calculating a range of predictive customer attributes, to support
marketing teams with their customer strategies. These attributes include, churn likelihood, in-market windows,
life-time value, engagement, product preference, and channel preference.
- Customer Segmentation - Supporting the segmentation of a customer base in a data driven way. Including
behavioural-based segmentations, traditional non-supervised machine learning segmentation, as well as supervised
segmentation combining customer value with demographic attributes. Used for both retention and social
acquisition.
- Recommendation Systems - Online and offline recommendations systems to support everything from
displaying recommendations in emails, on webpages, and as part of applications. Covering industries as diverse
as fashion retail, building supplies, quick-service restaurants, and music. Recommendations differentiated by
optimising predictions to achieve specific business outcomes.
- Marketing Spend Optimisation - Supporting marketing spend through channel analytics, price sensitivity
analysis, as well as reinforcement learning based exploration/exploitation approaches
- Lead Scoring - Identifying leads and accounts with highest probability of conversion, and whom should
be focussed on accordingly. Key to the success of Lead Scoring is enabling a sales team to prioritise
recommended leads as easily as possible
Inventory Use Cases
- Inventory Optimisation - Helping businesses who hold stock know how much to hold, and how much to
order. Across a range of industries, business models, and personas; for a mix of core, seasonal, and promotional
products; accommodating stock building, fast-fashion style wide-then-deep buying, and supplier constraints like
Minimum Order Quantities.
- Forecasting - Improving forecasts, and helping businesses get more out of the ones they already have.
Includes validation of forecast performance, and consideration of important concepts like hierarchical
forecasting, seasonality, the cold-start problem, and cannibalisation.
- Allocation - Distribute inventory across locations (such as FCs and Stores) in an optimal way.
Specific examples include allocation of end-of-season stock across their European factory outlets for
leading fashion brands, or distribute fresh produce across a grocery network in order to minimise penalties.
- Supplier Optimisation - Determining the optimal mix suppliers to acquire inventory from, given arbitrary
business rules and constraints; given many-to-one, one-to-many, and many-to-many inventory-supplier
relationships.
- Distribution Resource Planning - Managing the distribution of inventory and resources across various
locations to optimize efficiency and meet demand, assuming a network of stock and a distribution capacity that
can be utilised.
Pricing Use Cases
- Price Sensitivity Modelling - Analysis of price sensitivity and its drivers, to act as decision support
for pricing teams.
- Promotions - Developing and optimizing promotional strategies to drive sales and customer engagement,
including for close-out. Effective promotion planning usually considers maximising a set of business KPIs like
margin, engagement, and sell-through, given constraints (such as the promotion window)
- Quote Price Optimisation - Optimizing pricing for quotes and proposals to improve win rates and
profitability. By dynamically adjusting quote prices based on customer value and market conditions,
businesses can increase sales effectiveness.
- List Price Optimisation - Using AI to optimise standard list prices, to balance revenue and market
competitiveness. Often requires exploration of the pricing surface through bandit style solutions.
Generative AI Use Cases
- Explainability - Bridging the gap between quantitative forms of model explainability, and natural
language (of the domain being considered).
- Natural Language Interfaces - Automation of business-supplied constraints to a model in a non-technical
way, granting business users more autonomy when using a solution, and in turn helping them move faster
Strategy Support
- AI Project Delivery - Ground up development and design of AI-centric project delivery framework.
Flexible enough to accommodate deliveries ranging from mostly-standard to fully-bespoke, and accounting for the
unique challenges faced in AI delivery around factors like Scope, Feasibility, Value, and Adoption.
- Strategic Insight - started and grew an insight analytics consulting team, with a focus on actionable
and commercially impactful insight. Turned this into a commercial offering that was so successful Peak changed
its broader sales strategy around it.
- Adoption and Change Management - Often, delivery of a powerful AI application is the easy part and
getting end users to trust it and adopt it's outputs is much harder! It is critical to start these conversations
early and to ensure end user workflows are appropriately understood when developing a solution. Taking
experience
of 80+ successful AI deliveries, started small AI Adoption CoE with responsibility for supporting on our most
challenging (usually enterprise) adoption cases.
- General AI Strategy - over the years, I have consulted a number of businesses on how to develop and execute an
AI strategy. Three of the most common challenges businesses have are:
- Identifying the write use cases for AI in your business
- How to recruit, retain, grow, and structure a mixed team of AI professionals
- Maximising commercial (or product) impact of data analysts, data scientists, and ML Engineers
Industry Partner Collaborations
- AWS - Supported launch of AWS ML Marketplace, by delivering the first 6 business-focussed models
- AWS - Support launch of AWS Forecast and Personalise products
- Intel - Support Intel with testing of new chip for acceleration of industry AI solutions
- MunichRe - Helped scoped, define, then bring what is possibly the worlds first "Inventory AI Gurantee" to market.
Open Source Software
- citrus - R package for customer segmentation. Available on CRAN
Research and Academic Collaboration
Academic Funding, Collaborations, and Teaching
- KTP Sponsorship / Supervision - University of Manchester - Using entity resolution to develop a more
systematic and cost-effective means of cleaning and integrating customer data into Peak's systems. Alex Bogatu
- PhD Sponsorship / Supervision - University of Manchester - Adaptive AI Decision Agents for Holistic
Supply Chain
Optimisation: Merging Societal and Business Objectives. Rifny Rachman
- PhD Sponsorship / Supervision - University of Liverpool - Ethnomethodology in the Development of AI
Products. Dipanjan Saha
- Funded Research Project - University of Manchester - Use of NLP for Keyword Generation and Multiarm
Bandits
for
Campaign Optimisation in Google Adwords
- Post-grad Student placement projects - University College London - Hosted several PhD student projects,
focused on Supply Chain Simulation and Optimisation
- MSc - 20+ MSc Placements. Assorted universities (Manchester, Bath, Lancaster, Edinburgh, UCL)
- Varied Support - Edge Hill University - including industry talks and review of course content
- Varied Teaching Assistant support - Griffith University - mostly first and second year physics and
maths
Academic Publications
- Cost-effective Variational Active Entity Resolution - A. Bogatu, N. W. Paton, M. Douthwaite, S. Davie
and
A. Freitas 2021 IEEE 37th International Conference on Data Engineering (ICDE), Chania, Greece, 2021, pp.
1272-1283, doi: 10.1109/ICDE51399.2021.00114.
- The accuracy of ab initio calculations without ab initio calculations for charged systems: Kriging
predictions of atomistic properties for ions in aqueous solutions - Di Pasquale, Nicodemo; Davie, Stuart
J;
Popelier, Paul LA: The Journal of Chemical Physics, 148(24), 241724, 2018
- The long-range convergence of the energetic properties of the water monomer in bulk water at room
temperature - Davie, Stuart J; Maxwell, Peter I; Popelier, Paul LA: Physical Chemistry Chemical Physics,
19(31), 20941--20948, 2017
- Geometry Optimization with Machine Trained Topological Atoms - Zielinski, Francois; Maxwell, Peter I;
Fletcher, Timothy L; Davie, Stuart J; Di Pasquale, Nicodemo; Cardamone, Salvatore; Mills, Matthew JL; Popelier,
Paul LA: Scientific reports, 7(1), 12817, 2017
- Optimization algorithms in optimal predictions of atomistic properties by kriging - Di Pasquale,
Nicodemo; Davie, Stuart J; Popelier, Paul LA: Journal of chemical theory and computation, 12(4),1499--1513,2016
- Incorporation of local structure into kriging models for the prediction of atomistic properties in the
water
decamer - Davie, Stuart J; Di Pasquale, Nicodemo; Popelier, Paul LA: Journal of computational chemistry,
37(27), 2409--2422, 2016
- Kriging atomic properties with a variable number of inputs - Davie, Stuart J; Di Pasquale, Nicodemo;
Popelier, Paul LA: The Journal of chemical physics, 145(10), 104104, 2016
- FEREBUS: Highly parallelized engine for kriging training - Di Pasquale, Nicodemo; Bane, Michael; Davie,
Stuart J; Popelier, Paul LA: Journal of computational chemistry, 37(29), 2606--2616, 2016
- Prediction of intramolecular polarization of aromatic amino acids using kriging machine learning -
Fletcher, Timothy L; Davie, Stuart J; Popelier, Paul LA: Journal of chemical theory and
computation,10(9),3708--3719,2014
- Applicability of optimal protocols and the Jarzynski equality - Davie, Stuart J; Jepps, Owen G;
Rondoni,
Lamberto; Reid, James C; Searles, Debra J: Physica Scripta, 89(4), 048002, 2014
- Free Energy Calculations with Reduced Potential Cutoff Radii - Davie, Stuart J; Reid, James C; Searles,
Debra J: Journal of chemical theory and computation, 9(4),2083--2089,2013
- Relative Free Energies from Non-Equilibrium Simulations: Application to Changes in Density - Davie,
Stuart James, 2013
- The free energy of expansion and contraction: treatment of arbitrary systems using the Jarzynski
equality
- Davie, Stuart J; Reid, James C; Searles, Debra J: The Journal of chemical physics, 136(17),174111,2012