ScalaUA-2020 Abstracts

Alexander Merritt – ONNX-Scala: An ONNX (Open Neural Network eXchange) API, Code Generator and Backend for Typeful, Numerically Generic, Functional Deep Learning in Scala
  The talk will open by providing a sketch of machine learning and ONNX/ONNX-Scala and motivating the use of an interchange format and a dedicated serving/prediction layer for production. This will be followed by a discussion of the type-safe/functional features offered by ONNX-Scala and some remarks on performance. Next we will briefly introduce at a high level an example machine learning model in the context of a specific use case. Finally, we will walk through a full pipeline deployment from training (in one of several deep learning frameworks) to prediction/serving (in ONNX-Scala) for this model.
Anatolii Kmetiuk – An argument against functional programming
  In this talk, I'd like to discuss immutability, purity, category theory and why you do not need these to write sane code.
Artur Skowronski – Landscape after a battle – what’s left of Blockchain tooling
  It is well known that during the gold rush, the best money gets shovel sellers. In our industry, it's a similar thing - during the "Blockchain Gold Rush" a lot of developer's tooling emerged to suit the demand. Now, when we can say that the dust finally settled, let's do a bit of post-mortem and check what is a current landscape of Blockchain tooling industry. Are there any "shovels" that stood the trial of time?
Avi Levi – Domain Driven Design In practice
  Domain-Driven Design is a topic that is buzzing around for a long time however a lot of architects/engineers are struggling and don't know where and how to start. In this talk, we will take a real-life example and we will show step by step how to implement the DDD approach using akka & scala.
Ayush Mittal – Effectful* programming in Scala
  Functional Programming (FP) languages are about "purity". A pure function is referentially transparent and has no side effects. However, any enterprise application software must perform side effects like read from database or upload to an Amazon S3 bucket. It appears that the only way to make a code that performs side-effects referentially transparent is to not run it! In this talk, we look at the principle of "first-class effects". They are effects that don’t break referential transparency and allow users to model side effects in a functional style.