DeveloperWeek Austin 2018

Automation and continuous delivery can do quite a bit more than just make deployments faster and easier.  Learn how we started from scratch and used automation, continuous integration & delivery, and disposable environments to help transform an engineering culture in about 60 days.  

Hear Prat Vemana, Chief Product and Experience Officer at The Home Depot, discuss how the world’s largest home improvement retailer uses agile and end-user focused methods to build experiences and solutions for customers and associates. With the pace of retail technology changing faster than ever and 90% of their software written in house, The Home Depot is creating interconnected solutions to meet the tailored needs of their customers both in stores an online.

How do you deploy and manage not just 1, but 100’s of machine learning models?  What changes when you move from batch processing to synchronous calls?  When your algorithms become mission critical, what changes?  What is seemingly easy when working for a small company, with lower volumes or with batch operations becomes exponentially more difficult at scale.  Off the shelf development and deployment platforms have existed for years, but use cases are becoming broader and corporations are starting to customize and invest internally.  In this presentation, I will discuss the infrastructure and processes necessary to operationalize data science at scale.  

In this era of technological advancements, the challenge in harvesting data is not as cumbersome as it used to be. With the collection of meaningful data becoming easier and more accessible, it’s often difficult to keep up by creating meaningful applications to analyze and present this data. In this session, we’ll unravel that challenge: how can we make meaningful decisions with the data we collect? How enable ourselves and our app users to find the hidden value in their data? The answer to these questions runs parallel to development in data analytics itself. We’ll discuss how through associative technology, developers can improve the effectiveness of the analysis they perform on data to achieve a deeper, more complete understanding.

Historically Integration was a step in a larger process of many IT projects.  Today, integration is the key to most Digital Transformation initiatives, providing the seamless connectivity among applications, APIs, and data sources both inside and outside your organization. Join us to learn an intelligent approach to integration and building services.  Discover how you can orchestrate complex activities with your legacy systems by learning from past transactions.  We will be covering innovative ways for Data Ingestion, Integration, Process orchestration, PMML, Spark, and Data lakes.

Neuroevolution is a subcategory of machine learning that applies evolutionary computation to generate artificial neural networks. Traditionally, neural networks are trained using the backpropagation algorithm. Back propagation is limited in its application, especially in scenarios where a training set of sufficient size is unavailable as in artificial life or evolutionary robotics. Backpropagation also puts a lot of constraints on the topology of the network. The number of hidden layers as well as the number of neurons per layer need to be known in advance, the activation function must be differentiable, the network must be fully connected and recurrent connections are not allowed. Neuroevolution provides an alternative approach which generates nets that don't suffer from these limitations and can evolve to any topology.


In this talk, I will take you on a journey where we first explore the foundation of artificial neural networks and evolutionary computation. In the latter half you will learn how to combine the two and create a general-purpose platform for neural network based systems from cleaning robots to financial oracles.

We all know that performance is a critical factor in the success of applications and websites, but the process for improving performance is not always clear. Code quality and infrastructure are of course important, but in many cases, you can make vast improvements to the end-user experience of your application by focusing on some very basic application delivery techniques. One such example is by implementing and optimizing caching in your application stack. Kevin Jones offers an introduction to the basic principles of content caching and outlines how NGINX caches responses, discussing how the cache metadata is stored on disk and NGINX’s unique management process. Along the way, Kevin shares techniques that can help both novice and advanced users see better performance from utilizing the web cache features included in NGINX, as he reviews and analyzes all of the caching directives and parameters available within NGINX, showcases a working example that enables basic caching functionality, and explores various advanced caching implementations from both an NGINX configuration and an architectural implementation perspective.

With the rise of DevOps, containers are at the brink of becoming a pervasive technology in Enterprise IT to accelerate application delivery for the business. When it comes to adopting containers in the enterprise, Security is the highest adoption barrier. Is your organization ready to address the security risks with containers for your DevOps environment? In this presentation, you'll learn about: - Best practices for addressing the top container security risks in a container environment including images, builds, registry, deployment, hosts, network, storage, APIs, monitoring & logging, and federation. - Automating and integrating security vulnerability management & compliance checking for container images in a DevOps CI/CD pipeline - Deployment strategies for deploying container security updates including recreate, rolling, blue/green, canary and a/b testing.

Developers have always been depended on  for their company's technology choices and product delivery.   Today, developers have more challenges than ever including rapidly evolving technologies, increased direct interaction with their clients and customers, and changing relationships with UX, Product and Program Management, and their own management chain.  We will explore how developers can help themselves navigate their new environment and discuss things such as very senior developer roles, potential moves to other disciplines, and when/if they should look at management roles.

Serverless is a powerful and rapidly growing architectural paradigm which abstracts infrastructure concerns such as deployment and scaling, allowing developers to focus more on creating code and solving business problems. In this session, we will start by discussing Serverless and its benefits, and then get more hands on as we dive into an overview of the Fn Project. The Fn Project is an open source, cloud-agnostic serverless functions platform that allows developers to run their own serverless infrastructure on their laptop, on prem, or in any cloud. We will discuss Fn's container-native approach to the functions platform, how it supports any language, and we'll look at its elegant and simple yet powerful design. Throughout, we will have some hands-on demos of Fn's development interface as well as some more advanced concepts such as workflows implemented with Fn Flow.

A modern infrastructure stack utilizes containers for packaging microservices, orchestration for service provisioning, and a service mesh for the application logic and control plane. Docker has won the “container wars.” Kubernetes has won the “orchestration wars.” But the service mesh war is just heating up. Istio has emerged as a promising de facto standard, but it’s a pluggable architecture and primary components are evolving. In this presentation, Rob Whiteley will discuss: How a service mesh provides granular control, security, and monitoring of inter-service communication. The three primary components -- Mixer, Pilot, and Istio-Auth -- and swappable alternatives. How to deploy enterprise-grade service mesh utilizing a sidecar proxy in the Istio environment.

Every growing company goes through managing technology challenges with scaling startups. How do you manage to build the right team, aligning business and technology strategies? How critical is the right attitude and alignment within engineering team and is that the only criteria for success? Should the technology stack be reconsidered with growth and scaling to switch to Microservices? Should ROI be considered for every change planned to address long-term goals? What processes are good enough to have enough visibility for commitments?

Come learn how to build the right team, align business and technology strategies while managing technology challenges with scaling startups.

Instead of treating data science projects as Ph.D dissertations, they need to be treated as what they are, entrepreneurial software projects. Build an MVM (Minimum Viable Model), get it in front of users and keep iterating on it. The implications of approaching data science from an entrepreneurial path are immediate and far reaching. I’ll briefly highlight three big points to keep in mind. - Build the Minimum Viable Model: The first model should be terrible, consider it a Minimum Viable Model. - Mitigate Risk with Funding Rounds: Risk mitigation through repeated project valuation. - Achieve Growth through Competition: Data science is evolving too fast for top down organizations to keep up. DataGig is perfectly poised to provide the entrepreneurial way described here. We push up the likelihood of success on individual projects by enabling a lean model. Data science has produced some of the greatest tech innovations in the past decade, but as practiced in many organizations it’s also completely unsustainable. Producing more relevant models, mitigating risk and keeping up with the pace of the field will require organizations to rethink how they do data science. Structuring data science as entrepreneurship makes it a great investment once again.

API are not just an integration technology performing some complicated actions in the background. Even though it doesn’t do something you can touch and feel, it is a product. API have capabilities, features, and instructions. Things that differentiate it, and bring value. They have needs that are critical to its success - just like any other product. The success of your Product(API) is determined by the pace developers are adopting it. If the word “API” is in your landscape as the answer to growing your organization’s profitability, this session is dedicated to helping you. In this talk you’ll understand the importance for bringing Developer Experience to the start of your product development, and bringing developers to the front of decision making. We’ll highlight the “Dos and Don’ts” of adopting a successful API strategy at large scale Enterprises. Finally, we’ll review successful ways of talking about APIs as Products.

The complexity and amount of data rises. Modern graph databases are designed to handle the complexity but still not for the amount of data. When hitting a certain size of a graph, many dedicated graph databases reach their limits in vertical or, most common, horizontal scalability. In this talk I’ll provide a brief overview about current approaches and their limits towards scalability. Dealing with complex data in a complex system doesn’t make things easier… but more fun finding a solution. Join me on my journey to handle billions of edges in a graph database.

Engineers are often too focused on developing products and not focused enough on developing themselves. This talk provides an executive viewpoint on what makes a good engineer, how to advance your career in its various seasons, and how to show the world how good you are at doing what you love. Learn how to avoid common pitfalls, how to set and reach new goals, and how to become recognized as a leader by your peers and in your industry.

In the next 10 years, we will live in a world where robots are embedded into our daily lives as our friends, teammates, and even as part of our families - performing helpful tasks, providing safety, and interacting with humans in friendly ways only currently known to science fiction. Misty is developing an easy-to-program robot to "democratize" robots - move them out of the world of "roboticist" and into the world of "Javascript and Python" developers so they can invent thousands of useful skills for everyday life.

Imagine a world where a software engineer is truly responsible for the code they write; from the development, testing, deployment, to post-production monitoring. Maintaining and curating an integration test suite takes a lot of work, even for an experienced quality assurance (QA) team. Airbnb decided to deviate from the norm of employing designated QA team, and instead built a system to prioritize end-to-end ownership of features by product engineers. Engineers are responsible for all of these things, including having to be “at the ready” 24/7/365 for any problem associated with their feature or product. In Airbnb’s case, this is the preferred initiative because in most cases, the engineer making changes to the product has the most incentive and context to investigate, or triage a failure. Without a QA team, the engineering team at Airbnb has been able to organically build a test suite that has caught hundreds of regressions per month. In this session, we will share best practices and pitfalls we learned along the way, including: 1. Owning your own impact and the engineering culture at Airbnb. 2. An overview of our test infrastructure and what powers this system behind the scenes. 3. Features of our in-house test harness, and how it helped us scale it for an engineering organization of 1000+ engineers. 4. Determine what matters, how to get the most your tests with the least amount of work.

Volumetric capture and photogrammetry bridge the gap between the real world and the virtual world. Through the use of advanced computer vision, machine learning, and mathematics to create realistic environments, objects, and people allowing higher engagment and immersion for users into a digital world. Within the last several years the considerable number of advances occurring in the areas of object recognition, tracking, scanning, and the computational abilities of the computers themselves that capturing is well within the reach of most developers. These breakthroughs in technology have not only increased viability of the platforms but also moved these assets out of the uncanny valley to the world of realism. Current techniques of communication with video, Augmented Reality (AR), and Virtual Reality (VR) fall short in information exchange and often require additional resources to assist in conveying the message. This all cost more time and money; ultimately reducing platform effectiveness. Volumetric capture and photogrammetry can be created in a couple of different ways but the more common uses the technique of capturing images from cameras and sensors to create 3D meshes which can be merged seamlessly into game engines, VR headsets, AR environments and MR worlds for deep psychological influence on users. Anywhere that a CG asset could be used traditionally would be a great opportunity to us a volumetric object for increased immersion. This talk will compare these techniques regarding technical specifications including package sizes, capture options, computing needs, cost analysis, as well as benefit and complexity of each style and their use cases for education. We will cover the engagement and retention in creating realism for digital formats and leveraging these tools.