disconnect.me.uk
š Phil Stubbings
š¤ Things
- šØ Oil painting / Photos Instagram
- šļø Random Blogging
- š» Coding Github
- š„ Kaggle
- š¦ Twitter
āØļø Work
Iām a Data Scientist and Software Engineer working in the areas of Artificial Intelligence and Machine Learning. I currently work in the area of [Privacy-enhancing technologies](https://en.wikipedia.org/wiki/Privacy-enhancing_technologies). I like to work at the intersection of Software Engineering and Research, building systems to solve complex and novel problems. CV on LinkedIn
š¢ Places
- Current - Personal endeavours <3
- Evogro Staff Software Engineer ā Edge computing/IoT.
- Wunderman Thompson MAP Staff Data Scientist ā Data privacy and privacy preserving analytics.
- Harvard University OpenDP Visiting Fellow ā Distributed training of neural networks with differential privacy.
- LiveRamp Staff Data Scientist ā Acquired DataFleets; Privacy preserving data analytics.
- DataFleets Research Software Engineer ā Data Privacy Startup; Algorithms for privacy-preserving machine learning.
- Office for National Statistics Lead Data Scientist ā .gov; Computer vision, optimisation, building novel economic indicators.
- Evogro Machine Learning Software Engineer ā AgriTech Startup; Computer vision and optimisation algorithms.
- Arkasoft Principal Software Engineer ā Cloud computing; Industrial optimisation, research and development.
- Velti Senior Software Engineer ā Distributed computing; High-throughput and scalable messaging systems.
- British Telecom Research Scientist ā Operations Research; Supply chain optimisation, workforce scheduling and optimisation.
šØāš Academics
- MSc Computer Science
- BSc Artificial Intelligence
- NVQ Fine Art
š Publications
Google Scholar profile
- Feature space hijacking attacks against differentially private split learning
- A hierarchical urban forest index using street-level imagery and deep learning
- Modular neural networks for recursive collaborative forecasting in the service chain
- Collaborative demand forecasting in service chains
š” Patents
- Differentially private split vertical learning
- Data analytics privacy platform with quantified re-identification risk
- Analytics platform for federated private data
š¤ Contact
- phil<2^13>@gmail.com