My Role

Project Manager

My Responsibilities

  • Main contact
  • Deliver campus
  • Continuous development together with customers, setting the right expectations
  • Co-create and communicate strategic vision with customer
  • Feedback to co-workers
  • Developed alumni platform by strategic pairing with the Combient initiative Twin
  • Mentors & Lecturers: Created framework for mentoring, onboarding, feedback, supporting
  • Eliminate blockers for CodeHub, mentors and lecturers
  • Responsible for learning platform being up to date and working properly
  • Recording and uploading of video material for campus
  • Account for participants time zone differences
  • Being the face of campus

Background

We were asked to develop courses for the Combient Walenberg companies. The courses are enabling the companies to Strengthening companies to be Data Driven from the inside. The campus is a highly interactive course where participants work hands on with their company data. The participants get to work on a company specific use case with the help and guidance from our mentors. The course is given online on our course platform and have participants from all over the world. The course contains: instruction videos, self study literature, knowledge assessment quiz, reflections on how to apply this to your job and weekly assignments. On top of this there are weekly live lectures and mentor meetings. The use cases the participants are working on are real use cases from their company. The goal is to Strengthening companies to be data-driven organizations Below are a screenshot of the platform: 2021-02-28_17-11-32

Detailed Course Content

Data Science Tools

Covering Introduction to our inhouse Jupyter Notebook hub called CodeHub. Introduction to Python, Numpy, Matplotlib & Pandas

Introduction to Data Science & Inspirational Use Case

What is Data Science, Overview of Data Science Landscape. Data cleaning, CRISP-DM Model

Supervised Learning

Examples of supervised learning, Regression, Cost functions, Classification, Accuracy/Precision/Recall, Decision Tree

Measuring Model Performance

Mean Squared Error, Root Mean Squared Error, R^2, Mean Absolute Error, Anomaly detection, Accuracy, Confusion Matrix, ROC/AUC/AUCROC, F1-metric, Model selection, Cross validation

Applying DS in Practice

Bias, Variance tradeoff, Overfitting, Feature Selection, Regularization, Feature Scaling, Feature Encoding, Data Exploration, Heatmap exploration

Unsupervised Learning

KMeans clustering, The elbow method, Dimensionality reduction, PCA, Feature reduction

Models for Learning

Decision Tree based: Random Forests, XGBoost (eXtreme Gradient Boosted trees), K-Nearest Neighbours Algorithm, Neural Networks, Unsupervised learning: Clustering algorithms, Gaussian Mixture, Hierarchical Clustering

Hypothesis Testing

A/B testing, RCT (Randomized Controlled Trial), Hypothesis testing outline, Confidence intervals when performing hypothesis testing, Fisher’s exact test, central limit theorem. Using p-values, Thompson sampling method

Final Use Case

Last part of the course is a final presentation of the use cases. The participants have 15 min to present their findings and answer questions. Use Case Matrix: Document where participants record key information about their project and chosen approach Project Plan: Timeline of the project