Vlad Iliescu

Vlad Iliescu

Head of AI at Strongbytes, Microsoft MVP on AI

Iaşi, Romania

By day I work at Strongbytes as Head of AI, and by night I organize NDR, an annual Artificial Intelligence conference held in two of Romania’s largest cities - Iasi and Bucharest. By twilight I’m a Microsoft Most Valuable Professional on AI, and I blog about life, the universe and software development at https://vladiliescu.net

Area of Expertise

  • Information & Communications Technology


  • Machine Learning & AI
  • Azure
  • Azure Machine Learning
  • Azure Machine Learning Studio

Real-Time Machine Learning with Azure ML Online Endpoints

Are you using the most appropriate machine learning approach for your use case?

Some machine learning tasks can be completed in real-time, while others require batch processing. This talk will compare and contrast the pros and cons of both approaches, so that you can assess the differences between batch and real-time inference. You will also learn how to do real-time ML using the new Azure ML Endpoints, going from deploying a simple ML model to iterating the changes you need to make in order to get a production-ready service.

By the end of this talk, you will be able to determine whether your projects need batch or real-time ML, and know how to deploy a real-time machine learning model using Azure ML Endpoints.

Deploying a Machine Learning Model with Azure ML Pipelines

Azure Machine Learning pipelines are independently executable workflows of complete machine learning tasks. They are extremely flexible, and can be used for a large variety of scenarios, from simple offline scoring scenarios, to complex deep learning architectures.

During this talk Vlad will discuss the difference between online and offline scoring, the need for applying solid development principles such as SOLID in machine learning, and how pipelines help in that regard.

He will also show how to create a pipeline for deploying and running a simple model, and discuss possibilities for improvement.

Azure Automated ML for Fun and Profit: Kaggle Competitions

Automated machine learning is the process of automating some or all of the phases in a machine learning pipeline, such as data pre-processing, feature selection, algorithm selection, and hyper-parameter optimization. One advantage of these techniques is the empowerment of users, users that may or may not have data science expertise, allowing them to identify machine learning pipelines for their problems so that they achieve a high level of accuracy while at the same time minimizing the time spent on these problems.

During this talk Vlad will demo and go into more depth on Microsoft’s Automated Machine Learning library. You will learn how to automatically train predictive models, which features are deemed important and which features are excluded, and also how you can take a peek under the hood of the auto-trained model.

The model’s performance will be evaluated in an almost-real-world scenario, by competing in a live machine learning competition - Kaggle’s classic Titanic competition and seeing how to gradually improve results.

Vlad Iliescu

Head of AI at Strongbytes, Microsoft MVP on AI

Iaşi, Romania