Talks

How do we move information realtime and connect machine learning models to make decisions on our business data? This presentation goes through machine learning and Kafka tools that would help achieve that goal.

In the alternative JVM Language arena, Scala has come out with a new feature set under the code name "Dotty", now called Scala 3. Let’s take a look at some of the cool new stuff that we can use. This presentation will assume basic Java knowledge and no Scala knowledge is required.

Helm is excellent stuff. It is the package manager for Kubernetes and a way that you can package your Kubernetes YAML files and templatize them for reuse for either people in your organization or the public at large. Helm makes it super simple to deploy any major system with relative ease: find a repository, add the repository, and install your product. All Helm charts have a set of default values for everyday use cases, and it would be up to you to override the values you desire for your custom installation.

In reality, our jobs aren’t dealing with new code, but old crufty and evil code. How do we test such code, how do we get through it? How can we surgically take some of this bad code apart and make it testable.

One of the other essential tasks regarding the JVM is monitoring. How much stack and heap is your JVM using? What CPU saturation are you using? How many threads are being used, and what kind of threads are being used? What is your garbage collector looking like? How can I tap into the JVM and monitor other aspects of the JVM? All these questions are essential to ask since often administrators of your Java application will need to know these values to deploy and monitor your application.

Hey. Remember that time when we used to create jar or war files and we used to just ssh into a box and deploy on a single box? Well, it was simpler but also maybe that wasn’t that great of an idea. Time has certainly moved on, and our releases have become very advanced with very technical CI/CD pipelines, docker or debian packages, multi-purpose testing, producing signatures, perform security scans, perform releases, and then when you’re done tell the whole world about it. Whew! This presentation introduces JReleaser, a release platform for Java that does a multitude of chores for you.

Time is very precious and is often threatened by phone calls, emails, co-workers, bosses, and most of all, yourself. The Pomodoro Technique reigns in unfocused time and gives your work the urgency and the attention it needs, and it’s done with a kitchen timer.

JDK 11 saw the advent of a new HTTP Client and important new API for calling content on remote RESTful endpoints. This presentation will just focus on the HTTP Client are how to maximize its use.

Jupyter Lab has been a platform for Data Analysts and Data Scientists for the last few years. Still, it may expand to a more general population, including students, financial analysts, and other Scientific rigors. Running a Jupyter Lab today is just as important as running a web browser. It is an essential platform for learning, conveying information, and telling a story.

Property Based testing is letting the computer determine aggressive inputs for your subjects. When the computer provides edge cases, it highlights some of the typical problems that may still show up, so that you, the programmer, may have not thought about.

Join us for a session on MLOps, where we delve into the transformative practices and tools that bridge the gap between machine learning development and production deployment. Discover how MLOps enhances collaboration, reproducibility, and scalability in machine learning projects, ensuring seamless transitions from data engineering to model monitoring. Learn about the latest technologies, including Docker, Kubernetes, and MLflow, and explore real-world case studies highlighting best practices and common challenges. Whether you’re a data scientist, engineer, or manager, this session will equip you with the knowledge to streamline your ML workflows and drive impactful business outcomes.

This presentation will assume that the attendees have little to no knowledge of creating and operationalizing ML Models.

Imagine toString, equals, and hashCode in a single class. Can you change implementations on the spot? Probably not, there may be too many dependencies on your implementation. Time to break out an adapter pattern, a utility class, or better yet, a type class! A type class is a kind of template in very static functional programming languages. Imagine a template that can read, write information as a side effect as well? Type classes are powerful.

For those still grappling with Generics? This will be an attempt to clear the air about generics. What are wildcards? What is extends? What is super? What is covariance? What is contravariance? What is invariance? What is erasure? Why and when do I need this?

This presentation will discuss the patterns required when things go wrong in architecture and how to stay resilient under pressure. These are battle-tested patterns that every architect should know.

Hashicorp Vault stores encrypted secrets securely. You can store anything that you want into Vault including API keys, passwords, and certificates. Vault can also store dynamic secrets where it can negotiate with a cloud service on your behalf without direct interaction with your API keys. Hashicorp Vault is well thought out "bank" of information that handles storage, encryption, leasing, sealing.

Kafka is more than just a messaging pub/sub with storage. It goes beyond that and with technology from Confluent open source it has become a full-fledged data ETL and data streaming ecosystem.

In the past, testing the data layer was hard, really hard. It would include a test database, as opposed to the production which of course we didn’t want to use for testing. Maintaining any sort of consistency was a pain to do, and perhaps fell by the wayside because it was too tedious and too fragile. Then we used in-memory databases, but the only viable ones were HSQLDB or H2. The problem there though is that the SQL for the in-memory databases didn’t quite match up. Today we have Test Containers which uses a Docker Client to download a specific version of whatever database or messaging system you wish to integrate. The best part, you can test against the same version of the database or messaging system so you have higher confidence that your tests work.

Starting with JDK 5, we have had Futures, and they mostly went ignored. Now with concurrency and reactive technology in demand, it is essential that we understand what futures are, and how to handle them and make use of their power in asynchronous systems.

Facial Recognition whether you agree with it or not it is here to stay. Understand what the algebra is behind facial recognition, what is involved, and what kind of software there is out there.

Back in 2005, when an unknown program called Hudson came onto the scene, I was pretty excited. I had used CruiseControl up to that point and was getting pretty comfortable with the notion of Continuous Integration. Hudson had terrific graphics and an intuitive UI, unlike Cruise Control with XML laden directives. I thought the future was here. Then some cracks started showing, particularly in the delivery aspect of Hudson. I had used a Tomcat plugin, but trying to get a jar, use various environments, became a juggling act. Then came Docker, and I loved it because we get to deploy an image! No more SSHing into a box and messing things up. We had immutability. Now, we have Kubernetes, a system that takes these immutable containers and monitors them, among other great features. Jenkins still has its place but as CI, Continuous Integration. In this presentation, I will demonstrate Spinnaker, a system developed by Netflix for CD, Continuous Delivery. I will demonstrate how to set up, deploy, , monitor, rollback, and scale our pods with the ease of just selecting an item on a menu.

We take a look at another facet of architectural design: how we develop and maintain transactions in architecture. Here, we will discuss some common patterns for transactions: Outbox and Saga.

Kafka has captured mindshare in the data records streaming market, and in this presentation, we knock on its door and see what lies behind it. What is the draw? What makes it an attractive addition? How does it compare to Message Queues and other message streaming services?

Java 8 is pretty great, but mix in JavaSlang (now called Vavr) and get ready for some functional programming excitement.

Spark has a machine learning aspect to it and it’s called Spark MLLib. We discuss an intro into machine learning, some models, then apply some of those common machine learning models.

There is a new way of Threading, which means it is time to prepare. Project Loom has introduced Java Virtual Threads, which is now available in Java 21. Virtual Threads are small Threads meant to perform quick operations with the need to procure long-running OS threads, which can prove expensive. In this presentation, we will learn how to use these threads, what does it mean in relationship with the rest of the Java API, and what does it mean for third-party libraries.

Hopefully, we started moving on from Java 8. One of the great benefits of doing so, and there are many, is a module system. It is a controversial topic indeed, but I am hoping in this presentation to make some solid arguments that it is an essential part of our development.

This presentation we will discuss Kafka Connect. Kafka Connect is an open source project from Confluent to either take information from a data store as a Source and possibly dump information into another datastore, a Sink.

Jupyter Notebooks has been a platform for Data Analysts and Data Scientists for the last few years but it may be expanding to a more general population including students, financial analysts, and other Scientific rigors. Running a Jupyter Notebook today is just as important as running a web browser. It is an essential platform for learning, conveying information, and telling a story.

Java advances quickly. It is incredible how much incremental change accumulates over time. JDK 17 is now three years old, and we are at JDK 22 as of 2024. In this session, I will take some select JEPs (Java Enhancement Process) and demonstrate what they are and their use case. Then you can be ahead of the curve, and give you all the information you need to sell and demand the next generation Java for your work or opensource initiative.

In this presentation, we will introduce neural networks slowly. First, we will describe the process of learning machine learning. Then, we will discuss the tools typically involved with machine learning and neural networks. The core of this presentation is taking small steps to achieve a big goal: understanding a neural network. This presentation assumes that the audience knows nothing about the internals of machine learning.

This session will focus on data governance and making data available within your enterprise. Who owns the data, how do we obtain the data, and what does governance look like?

We have gone through a lot when configuring our computers with a language, with an editor, and maybe even setting up Git. For every project, though, we may need a different set of tools. In one, you may require Java 17 with a particular version of Kotlin. Another will require an npm of a specific version. You could use something like SDKMan for Java or a node version manager for Node. The problem is that you must install it and ensure you do it correctly for every project. The global installation won’t cut it; the more complex your project dependencies, the more tedious it will get. Nix provides consistent installation of tools so that it gets right every time and for every project. It manages all project dependencies so you can consistently build your project.

ReactiveX is a set of Reactive Extensions developed by Netflix, and is developed for various programming languages, like Java, Scala, and Clojure. ReactiveX overhauls the observable design pattern to achieve reactive goals. This presentation will solely focus on the Java version of ReactiveX, RXJava.

We have gone through a lot when configuring our computers with a language, with an editor, and maybe even setting up Git. For every project, though, we may need a different set of tools. In one, you may require Java 17 with a particular version of Kotlin. Another will require an npm of a specific version. You could use something like SDKMan for Java or a node version manager for Node. The problem is that you must install it and ensure you do it correctly for every project. The global installation won’t cut it; the more complex your project dependencies, the more tedious it will get. Nix provides consistent installation of tools so that it gets it right every time and for every project. It manages all project dependencies so you can consistently build your project.

Test-Driven Development is a test-first practice that drives the code through tests. It is a simple practice but not an easy practice.

The details are what gets you. This presentation introduces you to Test Driven Development, how to get started, what modern tools are available, and the challenges you will face.

We have been using JUnit and doing TDD for years, but you can take testing further. In this session, we will discuss some tools you absolutely need for testing your code outside of the regular stack you currently use.

In this session, we will discuss architectural concerns regarding security. How do microservices communicate securely with one another? What are some of the checklist items that you need?

Java advances quickly. It is incredible how much incremental change accumulates over time. JDK 17 is now three years old, and we are at JDK 22 as of 2024. In this session, I will take some select JEPs (Java Enhancement Process) and demonstrate what they are and their use case. Then you can be ahead of the curve, and give you all the information you need to sell and demand the next generation Java for your work or opensource initiative.

Graal is a VM and an awesome VM at that. Able to run a variety of languages and fast. The execution times can be impressive too. This VM can run anything, JavaScript, Python 3, Ruby, R, JVM-based languages like Java, Scala, Kotlin, and LLVM-based languages such as C and C++.

Join us for an in-depth exploration of cutting-edge messaging styles in your large domain.

A redux of the classic Gang of Four pattern, "The Interpreter Pattern." This pattern, once relegated as the useless one, has proven more than useless. With the advent of Java 21, it is now a pattern that demands focus in light of sealed classes and pattern matching.

There are multiple elements to Kubernetes where each component seems like a character in a book, pods, services, deployments, secrets, jobs, config maps, and more. In this presentation, we just focus on the security aspect of Kubernetes and the components involved. Particularly centered around RBAC and ServiceAccounts. What they are, what they do. We discuss etcd and secrets. We will also discuss other options for security in Kubernetes.

The silent and deadly competitor to React and Angular. Meet Elm. All functional and client language. Absolutely Clean.

Canary Deployments are the last ingredient of any Continuous Delivery or Continuous Deployment rollout. A canary deployment is a deployment strategy that releases an application or service incrementally to a subset of users. All infrastructure in a target environment is updated in small phases (e.g., 2%, 25%, 75%, 100%). This control makes a canary release the lowest risk-prone compared to all other deployment strategies, like the blue-green strategy. If you need to back out of a production deployment quickly without much disruption, then canary deployments may be an excellent practice to set up.

Spark Streaming is one of the few additions that are available with Spark that uses its internal architecture and creates a Streaming processing framework to process data in real-time.

Many have already seen what Reactive Streaming can do: RXJava, Akka Streams, Project Reactor. Now reactive streaming is a part of the canonical package for Java and now we can handle asynchronous pipelines with boundaries and make better well thought out applications