Speakers
No unicorns, no caticorns, just software development
Aarushi Kansal
@aarushikansalSenior Backend Engineer at Netlify
Building Machine Learning Pipelines for Computer Vision
Aarushi Kansal
Senior Backend Engineer at Netlify
Aarushi is a senior backend engineer currently working at Netlify. She codes primarily in Go and a bit of Python. Passionate about increasing diversity in tech, she spends a lot of her time mentoring underrepresented groups.
Building Machine Learning Pipelines for Computer Vision
Machine learning and computer vision, have been studied for decades, but only recently have begun creeping into industry. This means we need to start applying engineering principles to make models useful, scalable, and portable. This is an introduction into how to continuously train and deploy ML.
Machine learning and computer vision are no longer futuristic concepts or confined to a lab in some university. The industry is rapidly employing ML in products. However, implementing useful models is incredibly difficult because we don’t yet have many tried and tested standards or practices for applying traditional software engineering tools and methods. In this talk, we will dive into how to automate continuous delivery and training of models. We will discuss an example architecture set up for computer vision models, using KubeFlow and Kubernetes. This is intended to be an introductory talk into what is needed when deploying machine learning models, rather than focusing too much on specific technologies.
Adam Paszke
@apaszkeMain author of PyTorch
PyTorch: a modern flexible HPC environment
Adam Paszke
Main author of PyTorch
Adam is an author of PyTorch. He has recently graduated from a Master’s program in Computer Science at the University of Warsaw, but he has already worked with multiple organizations such as Facebook AI Research, Google and NVIDIA. He is currently finishing his second major in Mathematics. His general interests include programming languages, graph theory, numerical computing and machine learning.
PyTorch: a modern flexible HPC environment
Anyone involved in data analysis in Python knows NumPy, which is now a de-facto standard for all kinds of data processing in this language. However, it turns out that NumPy lacks many of the recent advancements that other scientific computing packages can provide — take accelerator (GPU) support, automatic differentiation or distributed computing utilities to name a few. PyTorch tries to fill that gap, by providing flexible and Pythonic abstractions for those concepts, while building upon the rest of the Python ecosystem. What’s more, there’s a whole community around the package, which has been steadily growing since its release in 2017, making it the most popular library used for machine learning research these days. This talk is meant to introduce the library while building on familiar concepts from NumPy, and show how PyTorch can be leveraged to express many interesting machine learning and HPC workloads
Alex Soto
@alexsotobSoftware Engineer at Red Hat
Kubernetes-Native Java with Quarkus
Chaos is taking over my Kubernetes cluster
Alex Soto
Software Engineer at Red Hat
Alex is a Director of Developer Experience at Red Hat. He is passionate about Java world, software automation and he believes in the open source software model.
Alex is the creator of NoSQLUnit project, member of JSR374 (Java API for JSON Processing) Expert Group, the co-author of Testing Java Microservices book for Manning and contributor of several open source projects. A Java Champion since 2017, international speaker and teacher at Salle URL University, he has talked about new testing techniques for microservices and continuous delivery in the 21st century.
Kubernetes-Native Java with Quarkus
Kubernetes is becoming the de-facto platform to deploy our application nowadays. But this movement also implies some changes in the way we code our applications. Before this change, we just developed a monolith application where everything was up and running up front, now we are breaking down this monolith into (micro)services architecture and everything is interconnected with the network. Although it might seem easy, done properly is not an easy as there are some challenges to address that was not in a monolith architecture. In this session, we’re going to start discussing what are these challenges (ie fault tolerance, service discovery, open tracing, or health checks) and demonstrate how they can be solved using Eclipse MicroProfile specification.
Come to this session to learn how to develop a successful Kubernetes Native application using Quarkus, a Java ecosystem way to develop cloud-first, container-native, serverless focused and Kubernetes optimized.
Chaos is taking over my Kubernetes cluster
Chaos Engineering is used in a distributed system to test integrally all the application by simulating error conditions within the system and observes how the application reacts to that errors. With all this information and analyzing it correctly, you can write applications more resilient to the failures. This talk will provide an introduction to the principles of Chaos Engineering, how to perform experiments, identify the weakness of the architecture and fix these problems.
Come to this session to learn different tools like Istio, Chaos Toolkit or Glooshot to run Chaos Engineering in Kubernetes and what strategies you can use to prevent chaos from taking over your system.
Alexis Duque
@alexis0duqueDirector of R&D at Rtone
AI at the edge with Tensorflow Lite to Design the Future of Vertical Farming
Alexis Duque
Director of R&D at Rtone
Alexis Duque is Director of research and development at Rtone.
He holds a Master of Engineering degree in Telecommunications and a PhD in Computer Science from the University of Lyon after a thesis in computer science on Visible Light Communications for the IoT.
He is actively involved in a collaborative research project on lightweight cryptography for the IoT, for which he is the technical lead for Rtone.
He regularly attends scientific and IT conferences to share his knowledge during talks about IoT, cybersecurity, and machine learning.
AI at the edge with Tensorflow Lite to Design the Future of Vertical Farming
In this presentation, we will demonstrate how to deploy Deep Learning algorithms on IoT devices using TensorFlow Lite. I will show how trending technologies like IoT, ML and Tensorflow can make the world better and explain our use case in vertical farming, show code snippets and make a short demo.
While Machine Learning is usually deployed in the cloud, lightweight versions of these algorithms that fit for constrained IoT systems such as microcontrollers are appearing. Using Machine Learning « at-the-edge » has indeed several advantages such as the reduction of network latency, it provides better privacy, and are working offline.
In this presentation, we will demonstrate how to deploy Deep Learning algorithms on IoT devices thanks to TensorFlow Lite.
We will see how to use it to design a smart vertical farming system able to predict and optimize the plant growth, at home or in developing countries where a reliable Internet connection still is missing.
Ana Lebrón Moreno
Site Reliability Engineer Manager at Enova International
SRE Role in a mature DevOps organization
Ana Lebrón Moreno
Site Reliability Engineer Manager at Enova International
Ana is the leader of the Site Reliability Engineering team at Enova International (fintech company in Chicago). She has been in the technology industry for the last 10 years in different roles. She started her career as a system engineer in the aerospace industry working in Germany and Spain. In 2010 Ana undertook an MBA, specializing in Technology Operations and since then she has worked at Enova International leading change management, problem management and site reliability engineering teams. Ana is the founder of Chicago Women in Tech Conference (Chiwitcon), and passionate about DevOps and Agile practices.
SRE Role in a mature DevOps organization
If your company is adopting a DevOps culture and you want to create or evolve your SRE team, this is your talk! Learn the different team, ownership and on-call responsibilities models you could use, as well as how to define a solid incident response process.
SRE role is one of the hottest and most trending roles in the technology industry, but implementing such a team of highly qualified engineers who needs to be a jack of all trades - master of none, build software but also understand infrastructure, application and network resiliency, and have the ability to constantly handle high-stress situations such as major incidents is NOT EASY. I have evolved Enova’s culture from a highly reactive culture (solving more than 3000 tickets per year) to a DevOps organization in which reliability is a shared responsibility and we do more than ~5000 releases a year. I will walk the audience through this personal journey by explaining how they can structure and evolve their own Site Reliability Team. Explain the pros/cons of each model, and how to create an Incident Response Model that is scalable for their organization.
Ana Valdivia
@ana_valdiData Scientist at Trilateral Research
When the AI hype comes to punish the opressed
Ana Valdivia
Data Scientist at Trilateral Research
Ana Valdivia has a PhD in Machine Learning and Natural Language Processing (University of Granada, 2019). She currently works at Trilateral Research as a Data Scientist, where she develops projects in cross-disciplinary teams to transform society into a better place. She collaborates in other research projects related to fairness, accountability, and transparency, as well as civic technologies for democratic participation and design justice. In 2018, she was awarded by the University of Chicago under the Data Science for Social Good Fellowship program, in which she collaborated with the Ministry of Education of El Salvador to identify factors affecting school dropouts.
When the AI hype comes to punish the opressed
Digital technologies have made the collection of tones of data possible in recent years. This datafication process has implied rendering into AI systems – which are fed with data – many aspects of the world that had not been quantified before. This has led our governments, companies, and start-ups into an AI-fueled utopia, where everybody wants to happily invest money on these technologies. But, what have been the consequences of this effect on working-class people?
Through this talk, Ana will present different stories about how AI hype has impacted on people's lives.
She will explain how tech communities can mitigate this impact through fairness, accountability, and design justice.
Andrew Betts
Web Developer and Developer Advocate for Fastly
Andrew Betts
Web Developer and Developer Advocate for Fastly
Andrew is a web developer and developer advocate for Fastly, working with developers across the world to help make the web faster, more secure, more reliable and easier to work with. He founded a web consultancy which was ultimately acquired by the Financial Times, led the team that created their pioneering HTML5 web app, and founded the FT’s Labs division. He has also been an elected member of the W3C Technical Architecture Group, a committee of nine people who guide the development of the World Wide Web.
Antonio Fernández Anta
@afdezantaResearch Professor at IMDEA Networks
Blockchains, Micro-ledgers and Other Creatures
Antonio Fernández Anta
Research Professor at IMDEA Networks
Dr. Antonio Fernández Anta is a Research Professor at IMDEA Networks. Previously he was a Full Professor at the Universidad Rey Juan Carlos (URJC) and was on the Faculty of the Universidad Politécnica de Madrid (UPM), where he received an award for his research productivity. He was a postdoc at MIT from 1995 to 1997, and spent sabbatical years at Bell Labs Murray Hill and MIT Media Lab. He has been awarded the Premio Nacional de Informática "Aritmel" in 2019 and is a Mercator Fellow of the SFB MAKI in Germany since 2018. He has more than 25 years of research experience, and more than 200 scientific publications. He was the Chair of the Steering Committee of DISC and has served in the TPC of numerous conferences and workshops. He received his M.Sc. and Ph.D. from the University of SW Louisiana in 1992 and 1994, respectively. He completed his undergraduate studies at the UPM, having received awards at the university and national level for his academic performance. He is a Senior Member of ACM and IEEE.
Blockchains, Micro-ledgers and Other Creatures
There is a big hype about blockchains and distributed ledger technologies (DTL), and their potential to reshape many aspects of society. Unfortunately, currently, these terms include many different aspects of these technologies in such a way that it is difficult to identify what is really needed and what is not in a particular set up. In this talk we try to analyze blockchains and DTL from a formalized and abstract point of view, separating these different aspects. We will especially explore scenarios in which the most convenient solution is having multiple ledgers, having lightweight micro-ledgers, or even ledgers that do not guarantee order.
Antón Rodríguez
@antonmryData Engineer at Inditex Group
Antón Rodríguez
Data Engineer at Inditex Group
Anton is a Software Engineer focused on Data Pipelines (Spark) and Event Streaming (Kafka Streams, Flink). Nowadays he works for Inditex Group (Zara, Massimo Dutti, etc.) as Data Engineer. In the past, he worked as a specialist in Deployment Pipelines, API Management and Advanced Orchestration in distributed systems. He enjoys building Data Pipelines with Java/Scala and deploying them to the Cloud or Kubernetes.
He co-organizes the Vigo and Coruña Java User Groups (VigoJUG & CoruñaJUG). He also likes to speak at technical conferences and contribute to open source projects.
Antonio Vilches
@avilchesPrincipal Software Engineer at Shapelets
Antonio Vilches
Principal Software Engineer at Shapelets
Antonio is one of the core software engineers at Shapelets, holding BSc and MSc in Computer Sciences from the University of Malaga. He is a software performance gangster. Thus, he went further and pursued a Ph.D. in Parallel Computing, where he developed parallel patterns for CPU-GPU systems. Antonio is the lead team developer of the company, always keen to share his knowledge and help others with their stuff.
Ara Pulido
@arapulidoDeveloper Relations at Datadog
Navigating the Sea of Kubernetes Development Tooling
Ara Pulido
Developer Relations at Datadog
Ara Pulido has been working on open source infrastructure and system engineering for the past 12 years. She worked for Canonical for 9 years, helping making Ubuntu a great developer and IoT platform. After that, she worked 2.5 years full time on Kubernetes projects. Today, she is a Developer Advocate at Datadog, a job that allows her to continue learning new technology and teaching it to others.
Navigating the Sea of Kubernetes Development Tooling
Your company has decided to start migrating to Kubernetes. Exciting! It is time to set up your development environment so you can quickly test your application changes in a real Kubernetes cluster. But, what should your development environment look like? Should you use minikube? kind? Firekube? Something completely different?
In this demo-led session, we will have a look at some of the development tools that help you locally deploy your Kubernetes application as part of your development process. We will show the differences between each tool from a technical and user experience point of view, so you can choose the ones that better fit your use case and company’s needs.
Arno Schots
Cloud Solutions Engineering Director
Arno Schots
Cloud Solutions Engineering Director at Oracle
Arno Schots is a Cloud Solution Engineering Director at Oracle, leading a team of 30+ Cloud Architects in France, Italy, Spain and Portugal. These teams are responsible for helping customers move to the cloud. Combining Enterprise Architecture background with hands-on engineering and development experience, he is passionate about sharing this knowledge with customers and technology audiences around the world. In his free time, he can be found outdoor doing sports like trail running, cycling and surfing. Last but not least, he is co-organiser of the great Jonthebeach event.
David G. Simmons
@davidgsIoTHead of Development Relations at QuestDB
Using Cross-measurement Math to Synthesize Sensor Data for Digital Twins
David G. Simmons
Head of Development Relations at QuestDB
David Simmons is the Head of Development Relations at QuestDB. He’s been passionate about IoT for nearly 15 years and helped to develop the very first IoT Developer Platform before “IoT” was even ‘a thing.’ He’s always had a thing about pushing the edge and seeing what happens. David has held numerous technical evangelist roles at companies such as DragonflyIoT, Riverbed Technologies, and Sun Microsystems.
Using Cross-measurement Math to Synthesize Sensor Data for Digital Twins
Sensor data is great and all, but what you do with it is what makes it useful. Being able to synthesize this data using complex math helps you make sense of your data and, ultimately, is what enables the building of Digital Twins.
IoT Data is streaming in from a variety of sensors in a wide range of formats, but it is often necessary to use that data in complicated calculations to present a more meaningful view of the systems being monitored. With weather and environmental data, for instance, it is often necessary to combine data from several different measurements to arrive at a result. Some of these calculations can be extremely complex.
In this talk I’ll show how the new Flux Data Query Language can be used to do cross-measurement math to arrive at a better, more accurate, view of incoming data that can be calculated and presented in real time. I’ll show how to build and execute complex mathematical calculations in real time for better data analysis.
David Rey
Chief Data Officer at Idealista
Data, is it oil or soil?
David Rey
Chief Data Officer at Idealista
David Rey, Chief Data Officer at Idealista, with more than 15 years of experience in tech & data and having undertaken many entrepreneurial projects. In the recent times as Chief Data Officer in some companies I have been focused in creating down-to-earth data teams (including the science part) and enjoying the creation and launch of a number of data strategy and data monetization projects. Regarding my formal training I have a MsC in Computer Science from the University of Oviedo, MBA from IE and Msc in Statistical Learning from UNED.
Data, is it oil or soil?
Maybe you have heard too many times data is the new oil, few people dare to question that data is the heart of the success factors of a company. But can this term be referred to oil or should it rather call it soil? I will talk about Idealista /data and how to grow a company from a seed dataset, and why we call our data indeed soil.
Félix López
@flopezluisSenior Software Engineering Manager at Eventbrite
Félix López
Senior Software Engineering Manager at Eventbrite
Félix López is a senior software engineering manager at Eventbrite, previously an engineering manager at Google, with more than 18 years of experience. During his career, he has worked on web development, video games, distributed systems and fin-tech companies. He holds a Research Master in Intelligent Systems (including neural networks, speech processing, data mining, etc.). He is interested in Distributed Systems, Machine Learning and psychology.
Jaroslaw Rzepecki
@JarekRzepeckiSenior Research Engineer at Microsoft
Towards more Human-like Video Game Agents
Jaroslaw Rzepecki
Senior Research Engineer at Microsoft
Jaroslaw Rzepecki is a Senior Research Engineer at Microsoft Research Lab in Cambridge, UK. He has a MSc in theoretical physics from Nicolaus Copernicus University and a PhD in astrophysics from Heidelberg University. After completion of his PhD he joined video games industry where he worked on two published titles (DiRT 2, Grid). Jaroslaw has also worked as a research engineer in computer graphics, investigating solutions to the global illumination problem in video games. Currently he is working in the Game Intelligence group at MSR where he combines his passion for engineering, video games and machine learning to make video games more fun for everybody.
Towards more Human-like Video Game Agents
After a brief introduction to (Deep) Reinforcement Learning (RL), we will talk about technology stack that allows us to run RL experiments on a large scale.
We will discuss what is needed in order to enable RL experimentation in any video game, challenges we have faced scaling up to 100+ video game instances running in parallel and how we have solved them.
Finally, we will show examples of trained agents performing tasks in video game environments and some of the debug tools we use during agent training.
Come to our talk if you would like to learn how we at Microsoft Research use the latest techniques in Reinforcement Learning to create Human-like video game agents.
Jinal Parikh
@Jins__pTechnology Analyst at Goldman Sachs
Jinal Parikh
Technology Analyst at Goldman Sachs
Jinal Parikh is currently a Technology Analyst. She loves handling scale and researching about distributed systems in her spare time. Having worked previously with Morgan Stanley and a few startups, she is also a Google Women Techmakers Scholar 2017, performing outreach activities to foster a local community that compels women to persevere in Tech.
Joerg Gablonsky
Technical Fellow at Boeing Research and Technology
Joerg Gablonsky
Technical Fellow at Boeing Research and Technology
Joerg is the Chair of the Boeing Enterprise High-Performance Computing Council, and the technical lead for numerical optimization inside Boeing’s Research and Technology Group. He spent part of his career in Boeing’s IT organization, helping to establish a centralized High-Performance Computing Service as well as the Digital Transformation Environment, transforming how Boeing develops software. Now back in Boeing Research and Technology where he started his career, he is developing mathematical optimization methods and exposing those methods via modern software technologies.
Jörg Schad
@joerg_schadHead of Engineering & Machine Learning at ArangoDB
The case for a common Graph-Based Metadata Layer for Machine Learning Platforms
Jörg Schad
Head of Engineering & Machine Learning at ArangoDB
Jörg Schad is Head of Engineering and Machine Learning at ArangoDB. In a previous life, he has worked on or built machine learning pipelines in healthcare, distributed systems at Mesosphere, and in-memory databases. He received his Ph.D. for research around distributed databases and data analytics. He’s a frequent speaker at meetups, international conferences, and lecture halls.
Jörg Schad is Head of Machine Learning at ArangoDB. In a previous life, he has worked on or built container infrastructure and distributed systems at Mesosphere, and in-memory databases. He received his Ph.D. for research around distributed databases and data analytics. He’s a frequent speaker at meetups, international conferences, and lecture halls.
The case for a common Graph-Based Metadata Layer for Machine Learning Platforms
We all know data is important for Machine Learning, but as it turns out for operating Machine Learning Platforms Metadata is equally important. With the rapid and recent rise of data science, the Machine Learning Platforms being built are becoming more complex. For example, consider the various Kubeflow components: Distributed Training, Jupyter Notebooks, CI/CD, Hyperparameter Optimization, Feature store, and more. Each of these components is producing metadata: Different (versions) Datasets, different versions of Jupyter notebooks, different training parameters, test/training accuracy, different features, model serving statistics, and many more. For production use, it is critical to have a common view across all these metadata as we have to ask questions such as: Which Jupyter notebook has been used to build Model xyz currently running in production? If there is new data for a given dataset, which models (currently serving in production) have to be updated?
Juan Carlos Rico
@_JCRicoCloud Solutions Architect
Juan Carlos Rico
Cloud Solutions Architect at Oracle
Juan Carlos Rico is a Cloud Solutions Architect at Oracle, where he supports customers in Oracle’s Cloud adoption across EMEA. JC was born, and graduated in Computer Science, in Malaga, and joined the company in 2013 after some time working as a Software Engineer. Alongside technology, his main passion is spending time with his family, friends and love public speaking in international technology events like Oracle Open World or J On The Beach amongst others.
Lucas Bernardi
Principal Data Scientist at Booking.com
The 7 Powers of Machine Learning
Lucas Bernardi
Principal Data Scientist at Booking.com
Lucas Bernardi is Principal Data Scientist at Booking.com . His main focuses are Recommender Systems, Machine Learning Infrastructure and Causality & Machine Learning. He leads the Feedback Loops team and also contributes to the Experimentation framework. He blogs about data science and contributed to articles published in SIGIR, RecSys and KDD. He is an Engineer with background in Computer Science and Software Engineering.
The 7 Powers of Machine Learning
Machine Learning is an amazing technology. But, what can we actually do with it? How can we use it to build better products? In this talk I will introduce The 7 Powers of Machine Learning, a tool to help Product Teams to make the most out of this amazing technology, and showcase how we successfully used it in Booking.com.
Luis Vaquero
Global Head of Data Science at Dyson
Trees in your Contact Centre: Streamlining Text Analytics to Delight our Customers
Luis Vaquero
Global Head of Data Science at Dyson
Luis is the Global Head of Data Science at Dyson, delivering business value through the application of the art data techniques. Before that, Luis was a senior technical manager at HPE and HP Labs working in large data systems and consulting engagements for customers. Luis has led large teams distributed across geographies (Singapore, US, South America, EU/UK) and really hates gardening and planting trees.
Trees in your Contact Centre: Streamlining Text Analytics to Delight our Customers
This presentation will show a couple of use cases where ML techniques have been applied to improve the understanding of our customers. I will share how the process started and evolved and highlight common pitfalls that organisations venturing along the same journey may benefit from.
Łukasz Gebel
@rauluka7Software Engineer at TomTom
Do Developers Dream of Stateless Apps?
Łukasz Gebel
Software Engineer at TomTom
Łukasz Gebel: Software engineer at TomTom by day, machine learning enthusiast at night. My leading technology is Java and Java-based frameworks. On a daily basis, I work on designing, implementing and deploying distributed systems that work in cloud environments, such as Microsoft Azure and AWS. I’m interested in classification problems and multi-agent systems. I love to learn, read books and play football – in no particular order.
Do Developers Dream of Stateless Apps?
In Blade Runner by P. K. Dick, trained hunters had to retire problematic Androids. We, Developers, are similar to those hunters. Our job is to solve problems. State brings complexity and troubles. Getting rid of it is not always possible. How to make our stateful distributed system highly available?
It’s a story based on the experience that I gained while working on stateful distributed systems deployed in cloud environments (Azure, AWS). It includes what went well and what is more important, what went wrong. I’ll start with defining state and explain differences between stateful and stateless apps (it’s not so obvious!).
Then I’ll discuss the strategies that we can use in cloud environments to ensure high availability our or systems. We’ll go through scaling, multi-region deployments, and why sometimes we need to care where our machines are located.
In the third part of this talk, I’ll focus on tools that help us to deal with the state and their high availability features provided by cloud. I’ll show you the live demo of Azure SQL failover and compare it to Cosmos DB. I’ll also discuss Storage and Queues. Understanding the limitations of tools we use is as important as being aware of what happens under the hood. It is needed to build reliable architecture.
I’ll sum up the talk by explaining what is SLA and how to calculate it for your system (yes, there will be some math). So, are we problem hunters or we are haunted by problems? Join my presentation, make your system highly available and dream peaceful dreams.
Marta Rivera
@MartaRiveraAlbaLead Data Scientist at Clarity AI
Measuring Social Impact: domesticating big data streams with Airflow and Machine Learning
Marta Rivera
Lead Data Scientist at Clarity AI
Marta has always been interested in understanding the underlying mathematical basis of dynamical processes. During her career, she has had the privilege to focus on the discovery and understanding of the mathematical rules of nature. She studied Physics at the Universidad Autónoma at Madrid and finished a Masters and PhD in Biophysics in between Madrid and USA. During her PhD she developed mathematical models to test the optimality of the visual system of fruit flies. As a postdoctoral researcher, she worked in Spain, Portugal and USA on larval behaviour across species using mathematical modelling, machine learning and computer vision. Since 2016 she develops algorithms and mathematical models based on machine learning to unravel the mysteries of the market and to measure the social impact of companies towards a fairer world.
Measuring Social Impact: domesticating big data streams with Airflow and Machine Learning
In this talk, I will explain how we developed and continuously improve our Big Data pipeline to measure the social impact that can be attributed to every company or government in the world. In a giant effort to fully characterize social impact, we integrate aggregated transactional data from individual bank accounts, government budgets, company supplier disclosures, consumption surveys, product composition databases, United Nations consumption reports and worldwide industry to industry relationships together with proprietary cutting-edge ML algorithms. We take advantage of Airflow to integrate a heterogenic repertoire of python microservices that can be independently updated. This architecture allows us to keep improving the methodology implemented and adding extra data sources while continuously serving our clients.
Max Neunhoffer
@neunhoefSenior Developer & Architect at ArangoDB
The case for a common Graph-Based Metadata Layer for Machine Learning Platforms
Max Neunhoffer
Senior Developer & Architect at ArangoDB
Max Neunhöffer is a mathematician turned database developer. In his academic career he has worked for 16 years on the development and implementation of new algorithms in computer algebra. During this time he has juggled a lot with mathematical big data like group orbits containing trillions of points. Recently he has returned from St. Andrews to Germany, has shifted his focus to NoSQL databases, and now helps to develop ArangoDB. He has spoken at international conferences including Strata London or MesosCon Seattle.
The case for a common Graph-Based Metadata Layer for Machine Learning Platforms
We all know data is important for Machine Learning, but as it turns out for operating Machine Learning Platforms Metadata is equally important. With the rapid and recent rise of data science, the Machine Learning Platforms being built are becoming more complex. For example, consider the various Kubeflow components: Distributed Training, Jupyter Notebooks, CI/CD, Hyperparameter Optimization, Feature store, and more. Each of these components is producing metadata: Different (versions) Datasets, different versions of Jupyter notebooks, different training parameters, test/training accuracy, different features, model serving statistics, and many more. For production use, it is critical to have a common view across all these metadata as we have to ask questions such as: Which Jupyter notebook has been used to build Model xyz currently running in production? If there is new data for a given dataset, which models (currently serving in production) have to be updated?
Miro Cupak
@mirocupakCo-founder and VP Engineering at DNAstack
What's new in concurrency: threads and fibers for everyday Java developer
Miro Cupak
Co-founder and VP Engineering at DNAstack
Miro is a Co-founder and VP Engineering at DNAstack, where he builds a leading genomics cloud platform. He is a Java enthusiast with expertise in distributed systems and middleware, passionate about genetics and making meaningful software. Miro is the creator of the largest search and discovery engine of human genetic data and the author of a book on parallelization of genomic queries. In his spare time, he blogs and contributes to several open-source projects.
What's new in concurrency: threads and fibers for everyday Java developer
In the last couple of years, over 20 years after the introduction of threads, the Java community has been reimagining what the future of concurrency should look like. Project Loom is introducing support for high-throughput and lightweight parallelization to help developers write scalable software with better performance and lower footprint.
In this live-coding session, we explore the concurrency model of Java, and talk about the recent updates, as well as the future of the platform through the upcoming Project Loom. In particular, we focus on APIs that can be leveraged by our applications directly, and talk about how the new constructs affect the way we write software in the context of the current trends, such as reactive programming. Come and learn what’s new in the JDK, and how you effectively parallelize your software!
Nadieh Bremer
@NadiehBremerFreelance Data Visualization Designer
Visualizing Connections
Nadieh Bremer
Freelance Data Visualization Designer at Visual Cinnamon
Nadieh Bremer is a graduated Astronomer, turned data scientist, turned data visualization designer, based near Amsterdam. She's working as a freelancer under the name "Visual Cinnamon". As 2017's "Best Individual" in the Information is Beautiful Awards, she focuses on uniquely crafted (interactive) data visualizations that both engage and enlighten its audience. Working for companies such as Google News Lab & UNESCO to small start-ups. From printed magazines to interactive experiences online to more promotionally focused visuals for press releases, data-driven reports, and data art.
Visualizing Connections
Connections are a part of us, of the world. From the connections between people, between cultures, within language, and more. In these days when more data is collected daily than we could ever hope to explore, the variety in connections being gathered is opening up the possibility to visualize these (often complex) networks.
During this talk, Nadieh will take you through the design process of several of her (interactive) data visualization works, from personal projects to client work. The common thread they all share is that they all reveal connections, but all differently. From a family tree of 3000 people connected to the European Royal houses to those existing between our Intangible Cultural Heritage created for UNESCO, to connections we have drawn in the night skies and more. Revealing that all types connections are unique and revealing the intricacies that lie within them requires a creative, iterative and custom approach.
Natan Silnitsky
@NSilnitskyBackend infra engineer at Wix.com
Greyhound - Powerful Pure Functional Kafka library
Natan Silnitsky
Backend infra engineer at Wix.com
Natan is on the Data streaming team in charge of building event-driven libraries and tools on top of Kafka. Before that, he was part of a task force that was responsible for building the next generation CI system at Wix on top of Google’s Bazel build tool.
Has many years of experience as a developer of large scale web services - First in .Net, later in Scala.
Natan’s passions include clean code, dev velocity and great software design.
Greyhound - Powerful Pure Functional Kafka library
Introducing Greyhound - open-source Kafka client SDK wrapper. Greyhound harnesses Scala + ZIO functional, async and concurrency features. It offers rich functionality and cleaner, simpler APIs Greyhound is used by Wix developers in more than 1000 event-driven microservices.
Wix has finally released to open-source its Kafka client SDK wrapper called Greyhound. Completely re-written using the Scala functional library ZIO. Greyhound harnesses ZIO’s sophisticated async and concurrency features together with its easy composability to provide a superior experience to Kafka’s own client SDKs It offers rich functionality including - Trivial setup of message processing parallelisation, - Various fault-tolerant retry policies (for consumers AND producers), - Easy plug-ability of metrics publishing and context propagation and much more. This talk will also show how Greyhound is used by Wix developers in more than 1000 event-driven microservices.
Nicolas Fränkel
@nicolas_frankelDeveloper Advocate at Hazelcast
Introduction to data streaming
Nicolas Fränkel
Developer Advocate at Hazelcast
Developer Advocate with 15+ years' experience consulting for many different customers, in a wide range of contexts (such as telecoms, banking, insurances, large retail and public sector). Usually working on Java/Java EE and Spring technologies, but with focused interests like Rich Internet Applications, Testing, CI/CD and DevOps. Currently working for Hazelcast. Also double as a teacher in universities and higher education schools, a trainer and triples as a book author.
Introduction to data streaming
While “software is eating the world”, those who are able to best manage the huge mass of data will emerge out on the top.
The batch processing model has been faithfully serving us for decades. However, it might have reached the end of its usefulness for all but some very specific use-cases. As the pace of businesses increases, most of the time, decision-makers prefer slightly wrong data sooner, than 100% accurate data later. Stream processing - or data streaming - exactly matches this usage: instead of managing the entire bulk of data, manage pieces of them as soon as they become available.
In this talk, I’ll define the context in which the old batch processing model was born, the reasons that are behind the new stream processing one, how they compare, what are their pros and cons, and a list of existing technologies implementing the latter with their most prominent characteristics. I’ll conclude by describing in detail one possible use-case of data streaming that is not possible with batches: display in (near) real-time all trains in Switzerland and their position on a map. I’ll go through all the requirements and the design. Finally, using an OpenData endpoint and the Hazelcast platform, I’ll try to impress attendees with a working demo implementation of it.
Nuno Preguiça
@nunopreguicaAssociate Professor at DI FCT NOVA
Reconciling availability and safety in distributed databases
Nuno Preguiça
Associate Professor at DI FCT NOVA
Nuno Preguiça is Associate Professor at DI FCT NOVA, and leads the Computer Systems group of the NOVA LINCS research lab. The broad aim of his research is to allow efficient and correct data sharing among geo-distributed users. He has participated in a number of national and EU projects. He co-invented CRDTs and received a Google Research Award in 2009 for his work on solutions for cloud data management.
Reconciling availability and safety in distributed databases
By the CAP Theorem, a distributed database can ensure either Consistency under Partition (CP) or Availability under Partition (AP), but not both. This has led to a split in the database world, with CP databases providing strong consistency in one extreme, and AP databases favouring high availability in the other extreme. This talk will show that there is a whole spectrum of alternatives between these extremes, which can address the specific requirements of a large number of applications by providing the necessary consistency with good availability and low latency. Furthermore, we will discuss how AntidoteDB database explores this spectrum of alternatives.
Patrick Debois
@patrickdeboisDirector of Dev❤️Ops Relations at Snyk.io
How secure is your build/server?
Patrick Debois
Director of Dev❤️Ops Relations at Snyk.io
In order to understand current IT organizations, Patrick has taken a habit of changing both his consultancy role and the domain which he works in: sometimes as a developer, manager, sysadmin, tester and even as the customer.
He first presented concepts on Agile Infrastructure at Agile 2008 in Toronto, and in 2009 he organized the first devopsdays. Since then he has been promoting the notion of ‘devops’ to exchange ideas between these groups and show how they can help each other to achieve better results in business.
How secure is your build/server?
Development has changed over the years, from doing everything yourself to a 3rd party package for every function. Operations has changed too, running your own servers is now considered an exception. To the cloud! We have learned that we need to trust others, but as our parents used to say - don’t trust strangers. So we secure our production server more than ever.
Yet, in the middle sits this no man's land: “the build server”. We think it’s time to take a closer look at some of the good practices around securing builds & artifacts to improve our day to day level of trust.
With Marked Sherman statement “Development is now assembly” in mind, the talk will focus more on the package/artifact/repository aspect. Less on the app security inside the code itself or at the OS/Machine level.
This talk I will go into detail on:
- How to verify trust of your dependencies: from metadata, binaries and repositories
- How to provide trust to others that build upon your software
- How this ties into the concept of “reproducible builds”
- How a practical “Software Bill of Material” looks
- How the concepts of the “The Update Framework” (TUF) relate
- How you can implement secure packaging policies
It will explain these topics using practical/code examples from the Nodejs and Docker ecosystems. All this will be presented from the different viewpoints from “dev”, “sec” and “ops”.
Let’s take ownership of your trust, we are already responsible when things go wrong anyway.
Peter Van Hardenberg
@pvhComputer Science Researcher at Ink & Switch
Local-first software
Peter Van Hardenberg
Computer Science Researcher at Ink & Switch
Peter van Hardenberg is a computer science researcher working at the Ink & Switch industrial research lab exploring local-first software. Born in Canada and based in San Francisco, in past lives he's been a cloud infrastructure developer (at Heroku), written video games (for Nintendo DS), and studied climate change (as an arctic oceanographer.)
Local-first software
We've been conditioned to accept that the software we write only runs if we pay our Amazon bill and that the software we rely on can disappear one day because someone else didn't. We've lowered our expectations, too. Our software can only be so fast when it runs on a computer on the other side of the world.
It doesn't have to be like this. We can have the benefits of the cloud with fewer of the disadvantages by thinking local. In this talk we'll look at the research Ink & Switch has been conducting into local-first software: software that prioritizes your experience on your computer. We've combined web technologies, recent breakthroughs in computer science, and peer-to-peer data distribution to show how you can build software with real-time collaboration at every level that never goes offline because it runs on your computer. Best of all, by reducing the incidental complexity in development, we believe we can enable developers everywhere to accomplish more with less.
Philip Brisk
Researcher Professor at University of California, Riverside
Philip Brisk
Researcher Professor at University of California, Riverside
Philip Brisk received the B.S., M.S., and Ph.D. Degrees, all in Computer Science, from the University of California, Los Angeles (UCLA) in 2002, 2003, and 2006 respectively. From 2006-2009 he was a postdoctoral researcher at EPFL in Switzerland. Since 2009, he has been with the Department of Computer Science and Engineering at the University of California, Riverside. His research interests include the application of computer engineering principles to biological instrumentation, FPGAs and reconfigurable computing, and efficient implementation of computer systems. He is a Senior Member of the ACM and the IEEE.
Reuben Bond
@reubenbondDev at Microsoft - Orleans Team
Writing High-performance Managed Code
Reuben Bond
Dev at Microsoft - Orleans Team
Reuben is a distributed systems enthusiast and developer on the Orleans team in Azure PlayFab at Microsoft. He first joined Microsoft in 2011 on the Azure Active Directory team and launched the multi-dimensional metrics system used by internal Microsoft services. Shortly after leaving Microsoft, and returning to Australia, he became involved with Orleans as an external contributor and soon found himself back in Redmond helping to simplify distributed systems development for all developers.
Writing High-performance Managed Code
High-performance software leads to reduced operational cost and happier users. Low-level optimizations for high-level languages with their attached runtimes can be daunting or even mystifying since there is often a lack of information on how to get started. This talk covers tips, tricks, and tools for producing high-performance code, with a focus on C# (.NET), but much of the advice also applies to Java and other managed languages. We will talk about lessons learned while optimizing the performance of Microsoft's distributed application framework, Orleans, as well as a high-performance serialization library. The talk will cover optimizations to keep the Garbage Collector at bay, use memory efficiently, and aid the JIT compiler in producing optimal code. Throughout, we will demonstrate useful tools for measuring performance and identifying bottlenecks and underutilization.
Robert Virding
@rvirdingPrincipal Language Expert at Erlang Solutions
The Erlang Ecosystem
Robert Virding
Principal Language Expert at Erlang Solutions
Robert Virding is one of the co-inventors of Erlang and was an early member of the Ericsson Computer Science Lab. He took part in the original system design and contributed much of the original libraries, as well as to the current compiler.
He has always been interested in language design and implementation and at the lab he did a lot of work on the implementation of functional and logic languages. More recently he has done a number of implementations of different languages in Erlang, and on the Erlang system, which have been spread and used externally.
He is now Principal Language Expert at Erlang Solutions Ltd. and is regularly invited to teach and present throughout the world.
The Erlang Ecosystem
Erlang is in many ways quite old though many of the problems for which it used are quite modern. The Erlang language and system was designed around a set of requirements for telecom systems. They were distributed, massively concurrent systems which had to scale with demand, be capable of handling massive peak loads and never fail. The Erlang concurrency and error-handling model was developed around these requirements.
This talk will describe the development of the language and the design of systems based on the Erlang, look at the BEAM which is the Erlang VM, and how it supports the properties of the Erlang language and systems built on top of it. It will also look at further development with the introduction of new languages like Elixir and how they can be added into the Erlang environment - the Erlang ecosystem.
Roland Kuhn
@rolandkuhnCTO at Actyx
Roland Kuhn
CTO at Actyx
Dr. Roland Kuhn is CTO and co-founder of Actyx, a Munich-based company that makes state of the art software technology accessible to small and midsize factories. He also is the main author of Reactive Design Patterns and previously led the Akka team at Lightbend.
Roman Pickl
@rompicTechnical Project Manager at Elektrobit
Are we really moving faster? How visualizing flow changed the way we work
Roman Pickl
Technical Project Manager at Elektrobit
Roman is an experienced speaker at conferences, a technical project manager at Elektrobit and former CTO at fluidtime. Due to his background in software engineering, business administration and computer & electronics engineering, Roman has specialized in working at the crossroads of technology and business innovation, DevOps being the sweet spot.
Are we really moving faster? How visualizing flow changed the way we work
High workload. Putting in automation. Deploying new technologies. But are we really moving faster?
In this talk, we will look at how making work visible, using DevOps metrics, helped us to discover bottlenecks and improve flow.
Find out how we answered this fundamental question.
Are we really moving faster?
After putting in countless hours improving the deployment pipeline, investing in automation and deploying new technologies, it is time to ask this fundamental question: “Are we really moving faster?”
This is a story of how we made work visible by applying DevOps and Flow Metrics to discover bottlenecks and improve flow.
The session will leave you with concrete steps to implement key metrics, automatically collect and visualize them on an open-source dashboard and find an answer to this important question.
Key Takeaways
-
A Brief Intro to Value Stream Mapping
-
Actionable DevOps and Flow Metrics
-
An Implementation Example using an Open Source Solution
-
References and pointers to advanced material
Sergey Bykov
@sergeybykovPrincipal Software Development Lead at Microsoft
Sergey Bykov
Principal Software Development Lead at Microsoft
Joined Microsoft in 2001 and worked in several product groups, such as e-Business Servers, Embedded Devices, and Online Services, before moving to Research in 2008 to incubate Orleans. Sergey continues leading the Orleans team after open-sourcing the project, now within Microsoft Studios.
Tudor Gîrba
@girbaCo-Founder at Feenk.com
Moldable development
Tudor Gîrba
Co-Founder at Feenk.com
Tudor Gîrba (tudorgirba.com) is a software environmentalist and co-founder of feenk.com where he works with an amazing team on Glamorous Toolkit (gtoolkit.com), a novel IDE that enables moldable development (moldabledevelopment.com). In 2014, he won the prestigious Dahl-Nygaard Junior Prize for his software engineering research (aito.org), being the only recipient of that prize that was not a university professor.
Moldable development
Software is data. Data is shapeless. We, humans, need a shape to reason about anything. Tools provide the shape of data. So, we argue that tools are essential in software engineering and we show how controlling the shape of our tools can fundamentally change our perspective on software. Specifically, we argue that every problem in software can be exhibited in a way that a human can relate to. We introduce moldable development as a systematic discipline of creating custom tools to capture the context of software problems. We exemplify the message through concrete demos based on Glamorous Toolkit (gtoolkit.com).
Viktor Klang
@viktorklangDeputy CTO at Lightbend
CloudState: Serverless beyond Stateless
Viktor Klang
Deputy CTO at Lightbend
Viktor Klang, is a public speaker and programmer‚ specializing in distributed systems and concurrency. He is a prolific contributor to, and Emeritus Tech Lead of, the Akka project; co-founding member of CloudState.io, co-founding member of the Reactive Streams Special Interest Group; a contributor to the Scala Standard Library APIs, such as Future & Promise; and a Java Champion.
CloudState: Serverless beyond Stateless
Serverless, or pay-as-you-go computing, via Functions-as-a-Service (FaaS) has become a force to be reckoned with in modern Cloud Native Application Development. There is a big, glaring, problem though: FaaS is only for a subset of all use-cases—it’s intended for stateless transformation of data. Most business applications tend to have a significant need for stateful transformations and domain models, which has led to having to work around this limitation by embedding data access within functions, leading to high coupling, high latency, and contention under increased load.
In this session, we will demonstrate how domain entities offered as stateful services make Serverless applicable in a much broader scope than FaaS, using CloudState. Using techniques such as Event Sourcing, and scalable eventual consistency via Conflict-Free Replicated Data Types (CRDTs), CloudState offers a broad range of state management strategies.
CloudState.io is an open initiative, led by Lightbend, to solve the problems of the first generation of FaaS by creating an open standard and a reference implementation for a polyglot, polystore, polyAPI, serverless platform based on industry standards such as gRPC and Kubernetes, and state-of-the-art technology for Reactive Systems such as Akka.
Vinicius Senger
@vsengerSenior Technical Evangelist at AWS
Vinicius Senger
Senior Technical Evangelist at AWS
Vinicius Senger is a Senior Technical Evangelist at Amazon Web Services for Latin America and has worked with software development for the past 20+ years. Founder of Globalcode and The Developers Conference, Vinicius is considered a top 20 influencer in IoT development, having developed automation and robotics projects for cars, boats, houses, helmets and many other things. Since 2017 at AWS, he has been working with IoT, Serverless, Machine Learning, Artificial Intelligence, Alexa skills and traveled around the world talking about innovation with AWS technologies.
Vlad Mihalcea
@vlad_mihalceaCEO at Hypersistence
Vlad Mihalcea
CEO at Hypersistence
Vlad Mihalcea is the CEO of Hypersistence, and he also works as a Developer Advocate for the Hibernate project. He is passionate about enterprise systems, data access frameworks, and distributed systems.
He wrote hundreds of articles about Hibernate on vladmihalcea.com, and he has a gold badge for the Hibernate, Java, and JPA tags on StackOverflow. He’s also the author of High-Performance Java Persistence.
Without Boats
@withoutboatsResearcher on the Rust project
Asynchronous Programming in Rust
Without Boats
Researcher on the Rust project
Boats is a researcher who works on the Rust project, a new systems programming language. They are a member of the language design, and library teams, mostly focused on enabling users to write high-performance network services with Rust. They've been a contributor to Rust since 2014 and they've been leading the design of Rust's asynchronous IO features since 2018.
Asynchronous Programming in Rust
Rust is a new programming language designed to empower everyone to build reliable and efficient software. One major application of a language like Rust is to build very high-performance network services that can process large amounts of data or many user requests. These sorts of services need to use asynchronous, or nonblocking, IO to meet their performance requirements.
Rust has a rapidly growing async IO ecosystem and a built-in async/await syntax to support these use cases, but that wasn't always the case. This talk will trace the history of async IO in Rust, explore the trade-offs between different designs the project examined, explain the low-level details of the async/await system Rust has adopted, and give a look forward to the future of how Rust's async IO features will continue evolving.
Yara Senger
@yarasengerCEO at The Developer's Conference
What should developers learn about the Platform Revolution and Open Innovation?
Yara Senger
CEO at The Developer's Conference
Yara Senger is engaged in transforming the Brazilian IT ecosystem by empowering communities, helping professionals to achieve their career goals and build better software.
She is a Java Champion, JavaOne Rockstar, TEDx Speaker and also
the CEO of The Developer's Conference, the largest Developer's Conference in Latin America, engaging more than 100 local communities, thousands of speakers and more than 21.000 attendees per year.
She has created programs to attract more people to the Technology Industry and increase the diversity not only in the conferences but in the companies.
What should developers learn about the Platform Revolution and Open Innovation?
Developers are essential to driving digital transformation on many businesses and this is why executives must learn more about technology and developers should learn more about business.
During this talk we'll discuss 5 key points about Open Innovation and the Bussiness Platform Model:
What is the Business Platform Model and how it works
- What is open innovation and how open could it be?
- What are the most common launch strategies for platforms?
- Governance and policies for business platforms
- Why and How to bring open innovation to your business.