Daniel Nee Email & Phone Number

Stockholm, Stockholm County, Se

Stockholm, Stockholm County, Se

Blackfriars, England, Gb

DS Manager for the consumer product.Responsibilities:- Line management of 5 ML Data Scientists and 3 Product Data Scientists, embedded into product teams. Informal management of 3 Backend Engineers (1-1s, sprint planning).- Data Science Manager for 2 product teams - Merchandising and Consumer Pricing. Merchandising owns the main restaurant feed of the consumer product (incl. UI, navigation, ranking, recommendations & search). Consumer pricing owns all of consumer facing fees and price presentation.Tech stack: Python, Tensorflow, Snowflake, Looker and AWS

Welwyn Garden City, Hertfordshire, Gb

- Developed data products across 4 areas of the business; Store (space, range & display), Online Search & Recommendations, Price Optimisation and Forecasting.- Delivered an algorithm to optimise store space. Created sales & profit regression models and optimised store space using standard MILP solvers. Trial delivered +xx% sales growth and +xx% profit growth. Further rolled out to all large stores in 2018 (approx. £xxM profit). - Developed new grocery recommendation algorithms including a Search-To-Item recommendation which doubled Add-To-Basket rate from xx% to yy%. Techniques used included: collaborative filtering, markov chains and embeddings.- Line managed 9 ML Data Scientists across 2 locations. Helped grow the DS team to 32.- Created and established a data science performance management programme (levels & expectations, learning & development plan).- Built an academic relationship with UCL. Supervised 2 years of MSc research projects. Supervised ASI interns.Tech stack: Python, Scala, Hadoop, Spark, Hive, Oozie and AWS

Welwyn Garden City, Hertfordshire, Gb

- Developed and productionised data ingest of core data feeds into Hadoop (sales, clickstream, product catalogue, etc.). Used Spark to transform raw feeds into a performance optimised data layer used by over 150 Analysts and Data Scientists within Tesco. Drove support within the business to take Tesco from a 15 node POC cluster to 300 node production cluster and create a Hadoop data engineering team.- Created several data products: recommendations engine for Clubcard Boost, Clubcard Points promotional forecasting, share of spend model for targeted marketing and grocery churn prediction for targeted marketing.- Delivered various internal training courses (Introduction to Apache Spark, Statistical Testing). - Won best presentation at Tesco’s yearly internal data conference on multiple comparisons.- Recruited and managed a team of 3 ML Data Scientists.

London, Gb

- Developed new Hadoop based data processing and machine learning pipeline for creating marketing audience segments. Approx. 6,000 Logistic Regression models retrained daily and provided predictions across 200m cookies. Audience data produced was the primary revenue stream for the business.- Created algorithms for finding custom lookalike audiences. Re-framing existing solution to a classification problem. Delivered significantly lower cost per acquisitions (CPAs) for clients.- Used VisualDNA’s proprietary visual quiz to predict credit risk of banking customers. Evaluated and tested numerous different classification methods to tackle this small dataset problem, hugely prone to overfitting. Credit risk business area became revenue generating and was later purchased by CreditInfo.- Developed revenue and product funnel dashboards used in weekly exec team meetings.- Recruited and managed a team of 5 (2 Data Scientists, 1 Data Engineer and 2 Data Analysts).Tech stack: Scala, R, Hadoop, Scalding, Hive, Luigi and MySQL

Developed a ML approach for detecting ad viewability along with a content classifier that crawled and classified millions of websites into a hierarchy of content categories.Tech stack: R, C++, PHP and MySQL

Guildford, Surrey, Gb

Fraud detection for financial services clients, combining multiple disparate datasets using a mixture of statistical techniques and social network analysis.Tech stack: Java, SAS, R, SQL and NetReveal.

Southampton, Hampshire, Gb

Research Assistant in Information: Signals, Images, Systems group. Applying probabilistic and machine learning techniques to improve the base calling error rate in the Illumina DNA sequencing platform.Tech stack C++, Matlab

New York, Ny, Us

Internship as an application developer within the prime brokerage technology team. Working enhancements to production internal web-based tool for margin reporting. Tech stack: C#, ASP.NET, SQL Server and Crystal Reports.

Tool development for internal engineering team

Msc, Computational Statistics And Machine Learning

Bsc, Computer Science