Elisabeth Hendrickson

Elisabeth Hendrickson

Care and Feeding of Feedback Cycles


Nothing interrupts the continuous flow of value like bad surprises that require immediate attention: major defects; service outages; support escalations; or even scrapping just-completed capabilities that don’t actually meet business needs. You already know that the sooner you can discover a problem, the sooner and more smoothly you can remedy it. Agile practices involve testing early and often. However feedback cycles come in many forms. It’s not just about automated tests and CI. It’s also reading the mood in the room when pitching an idea or instrumenting the product to do cohort analysis in production. This talk examines the many forms of feedback, the questions each can answer, and the risks each can mitigate. We’ll look at what makes feedback lose value as well as fundamental principles that you can apply immediately to keep your feedback cycles healthy and happy.

Elisabeth Hendrickson, better known as @testobsessed, has been kicking around the software industry for a couple decades in a variety of roles including tester, developer, and agile enabler. Author of Explore It! from Pragmatic Books, she is also known for her Google Tech Talk on Agile Testing and popular Test Heuristics Cheatsheet. She won the prestigious Gordon Pask Award from the Agile Alliance in 2010. In 2012 after more than a decade as an independent consultant, she joined Pivotal, the company where she first learned eXtreme Programming (XP). She is now the VP R&D for Pivotal’s Big Data Suite.

Recent talks by Elisabeth include her XP 2016 keynote “XP At Scale” and her DevOps Enterprise Summit 2015 talk “It’s All About Feedback“. We’re all huge fans of Elisabeth’s work on eXtreme Programming, exploratory testing, and Continuous Delivery – and we’re looking forward to seeing her as a keynote speaker at PIPELINE 2018!

Read more about Elisabeth
John Allspaw

John Allspaw

Poised To Adapt: Continuous Delivery's Relationship To Resilience Engineering


Resilience Engineering is both a community of researchers as well as an interdisciplinary field of study. Continuous delivery can be seen as a source of what the RE community would call “sustained adaptive capacity.” I’d like to introduce the audience to what this means (in real terms, not academic) and hopefully open up new lines of dialogue and inquiry about how our “systems” work. It’s time to take human performance seriously in software, and this is a step in that direction

John Allspaw is the former chief technical officer (CTO) of Etsy, leading all of Etsy’s engineering efforts. Previously, he was the SVP of Infrastructure and Operations at Etsy. For over 20 years, he has worked in system operations in the biotech, government, and online media industries.

John began his career running vehicle crash simulations for the U.S. government. In 1997, he made the leap to the web, building the technical infrastructures at Salon, InfoWorld, Friendster and Flickr. John is the author of The Art of Capacity Planning and Web Operations: Keeping the Data on Time, both published by O’Reilly Media. He holds a BS from the University of Massachusetts, Amherst, and an MSc from Lund University in Human Factors and Systems Safety.”

Read more about John
Andrew Martin

Andrew Martin

Meteor-Proof Infrastructure: Reproducible Environments with Container Build Images (Audience: INTERMEDIATE)


Treating our infrastructure as immutable and expressing, versioning and storing it as code is now becoming common-practice. But how are changes to that code observed and tested? How do you know what code was used to build production infrastructure? Can you bring it back if it gets hit by a meteor? Don’t be a dinosaur, protect yourself from the inevitable with versioned, tested, and audited container build images.

Bake and version everything; manage secrets separately and securely; allow fast local iteration without impacting other users or environments; and recover from catastrophe, intrusion, or the dreaded fat finger of doom.”

Andrew has a strong test-first engineering background gained developing and deploying high volume web applications. Proficient in application development and systems architecture and maintenance, he is comfortable profiling and securing every tier of a bare metal or virtualised web stack, and has battle-hardened experience delivering containerised solutions to enterprise clients. He is a co-founder at https://control-plane.io

Read more about Andrew
Anna Shipman

Anna Shipman

How Coding In The Open can help you release faster (Audience: ALL)


The UK government is required to publish all new source code in the open. There are many benefits to coding in the open, and one of the interesting side-effects is that it makes continuous delivery much easier.

For example, when you code in the open, you tend to separate concerns better which supports continuous delivery. Open code makes collaboration easier, and the code being open means other people can see when the build fails and help fix it. Finally, it’s cheaper: GitHub and Travis are both free if your code is open. Not needing to purchase services can make the process of publishing your code much faster.

In this talk I will explain the ways that coding in the open supports continuous delivery. You will learn about the benefits and challenges of coding in the open, about how coding in the open can help you release faster, with examples of how this has worked in practice, and on the way you’ll get some insights into this transformation in the UK government”

Anna Shipman is Open Source Lead at the Government Digital Service. She is a technical architect and software developer. Prior to her Open Source role, she led the Government PaaS team, building a platform to host government services, and was a technical architect on GOV.UK. She is an international speaker. She blogstweets and is always up for a game of pool.

Read more about Anna
Chris Young

Chris Young

What can the ARPANET teach us about Continuous Delivery? (Audience: ALL)


Computer Networks are fundamental to Continuous Delivery. They enable the flow of delivery pipelines without interruption. With the prominence of public cloud computing the destination of that delivery is increasingly on the public Internet, even if it’s use is private to an organisation.

If we look back to the origin of computer networks and the Internet’s precursor, the ARPANET, we find a distributed system, which as it grew was updated remotely and re-engineered via what was arguably one of the first ever Continuous Delivery Pipelines.

What can we learn from those pioneering days that is relevant to our practice today?

In this talk we’ll explore how vision, purpose, monitoring, testing, remote updates, communities of practice and refactoring allowed the ARPANET and the Internet that followed to continuously deliver both the software that powered it and the increased value it provided as it grew.

There will be a fair amount of audience participation as we will model the growth of the ARPANET with some, not to scale, cardboard models and bits of string.

This talk draws heavily on Katie Hafner and Matthew Lyon’s book ‘Where wizards stay up late – The Origins of the Internet’”

Chris has been leading engineering teams since 1999. He was pivotal in the BBC’s adoption of Agile in 2000-2001. Since then he has worked for broadcasters and platform owners including Dish Network in the US and Deutsche Telekom in Europe. He is currently Vice President, Engineering at Linius Technologies. He is an active member of the Lean/Agile/DevOps community speaking at Meet Ups and conferences across Europe including GOTO Berlin, Agile Cambridge, CukeUp and QCon. His favourite language is Python.

Read more about Chris
Clare Macrae

Clare Macrae

Escaping 5 decades of monolithic annual releases (Audience: ALL)


“In this presentation, I will present lessons learned in moving from a large and monolithic 4-month long, annual release process for a software suite comprising numerous desktop applications and a large database.

Our context:
* a 52-year old database
* 3 million lines of C++
* One-year release cycle
* Very expensive manual testing
* Half our C++ developers are tied up for 4 months on one release!
* Result: critical bugs and high stress

A glimmer of hope:
* New test hardware, with efficient set-up process
* Automation of “smoke” tests
* Days of manual testing reduced to hours

Lessons learned – things to share:
* Forget abstract discussions about the benefits of automation
* Instead, empower teams to make incremental improvements
* And crucially, have them demo the benefits to internal stakeholders
* Use the pain of a difficult release to justify major improvements
* Just do it! Or just get started!
* And don’t wait for permission to make improvements!”

Clare has worked in technology for over 30 years, and is a Principal Scientific Software Engineer at Cambridge Crystallographic Data Centre. She currently leads the team that’s replacing their 30 year old database format. When not thinking about supporting teams to streamline development and releases, she spends time with family and friends, and more recently, also contributes C++ code to the CycleStreets journey planner.

Read more about Clare
Daniel Jones

Daniel Jones

Continuous Delivery is better for your brain (Audience: ALL)


Continuous Delivery doesn’t just feel better – it objectively is better. In this talk we’ll explore the psychology and neuroscience of continuously delivering on a platform, explaining how automation allows for greater sociability and empathy in teams, and how this is demonstrated to increase productivity. We’ll look at the scarcity effect, its negative impact sustainable velocity, and how continuously delivering the simplest thing can free us from it. Finally we’ll look at intrinsic motivation and how continuous delivery increases the frequency of reward, reducing the amount of technical debt a team incurs.

The OECD and Bank Of England have identified a productivity crisis – arguably caused in part by inefficient, outdated working practices. We need to change the way we work not just for the economic benefits, but for the human benefits too.”

Daniel Jones is CTO of UK Cloud Foundry consultancy EngineerBetter, who have improved value-delivery at the world’s largest banks, global wealth management firms, FTSE 100 retailers, governments and non-profit organisations. Daniel has is a regular speaker, and has also been an IT Consultant, an award-winning shorinji-kan jiu-jitsu instructor, indie games developer, entrepreneur, and also once stormed the stage at a music festival to expose bribe-taking by a hardcore punk band!

Read more about Daniel
Dave Farley

Dave Farley

Optimising Continuous Delivery (Audience: EXPERT)


The deployment pipeline is a central idea in Continuous Delivery. It represents the channel through which ideas inside your business flow to the outside world so that they can be delivered to your users in the form of working software. First steps in creating a working deployment pipeline are very rewarding. They help to improve the quality of the software that you create and usually lead to happier users.

As teams mature in their adoption of Continuous Delivery though, the pipeline often begins to slow. We have more tests, we are testing more complex things, our software becomes more complex and more difficult to deploy.

How do we address these growing pains? What are the next steps, beyond the CD beginner’s basic automation? How do we optimise our pipelines to meet our needs as these needs grow?

This talk will address the growing pains in Continuous Delivery and describe some strategies to cope with them.”

Dave Farley is the co-author of the Jolt-award winning book ‘Continuous Delivery’ a regular conference speaker and blogger and a contributor to the Reactive Manifesto. Dave is an independent software developer and consultant, and founder and director of Continuous Delivery Ltd.

Read more about Dave
Denise Yu

Denise Yu

What if you treat your CI pipeline as a product? (Audience: ALL)


There are many tools that facilitate a “CI as Code” approach to continuous delivery, but how many organisations are treating their CI infrastructure as an internally-facing product? Our software team has taken “CI as Code” quite literally over the last two years, as a dedicated team has developed a highly-configurable template for a Concourse pipeline. This template features opinionated APIs and useful abstractions that enable product teams to build, test, and ship open-sourced releases as well as artefacts for enterprise platform users. This talk will be equal parts experience report and cautionary tale: We’ll share the technical and cultural challenges of shared infrastructure templates from the perspectives of both producer and consumer, and we’ll make some recommendations about when and how you should consider implementing this pattern in your organisation. When your pipelines are code, it needs love just like everything else you ship!

Denise currently works as a full-time open source Cloud Foundry contributor on the Services API team, whose role is to define a standard specification for provisioning services on the platform. Prior to joining Pivotal she wrote web applications for a financial publishing company and worked at an educational technology startup. In her spare time, she enjoys traveling, playing classical guitar, and being a mediocre skier.

Read more about Denise
Elizabeth Ayer

Elizabeth Ayer

Beyond the Black Hole: Product Management for Continuous Delivery (Audience: ALL)


Continuous Delivery can be a hard sell to product managers, even in small, healthy organisations. The benefits sound implausible, the costs high, and it’s really hard to understand all that it enables.

However, CD has been hugely beneficial at the leading edge of product management. As deployment pipelines have taken root, there has also been a shift in how we see and measure user value. Take these trends together, and there’s been a giant leap forwards in how to prioritise work and judge its success.

When our teams at ORG stopped throwing features into a black hole, and instead closed the loop with feedback, it had surprising knock-on effects. Come to this session and find out more!”

Elizabeth Ayer started her career as a Java, C++, and ActionScript developer, but moved into product management after one too many dumb feature request. All her products have been for highly technical audiences, orchestrating engineering changes. She cut her teeth managing the enterprise platform under automotive, shipbuilding, and factory verticals at Siemens PLM (Product Lifecycle Management). Now she manages the portfolio at Redgate Software, a leader in Database DevOps tooling. In those 11 years she experienced the shift to Continuous Delivery for her own dev teams and in the processes of hundreds of customers using her products. The battle scars have left her intensely passionate about Agile and curious to push its boundaries.

Read more about Elizabeth
Kenny Baas

Kenny Baas

Disentangle the Data Enterprise Spaghetti to continuous delivery, a Domain Driven Design Story (Audience: INTERMEDIATE)


Are you working Agile but not doing twice the work in half the time? Does your Enterprise Architecture look like a spaghetti? Or is continuous delivery not working for you? Well, this is often the case at most enterprises… And, it is perfectly explainable why this happens.

Join me in this talk, about an enterprise trying to move faster with agile and continuous delivery, but actually finds itself slowed down and getting more entangled because of Conway’s Law. We will show how we used Event Storming to create a Context Map to get grip on the language and communication within the company. Then, we show by using the inverse of Conway’s Law how we disentangled the spaghetti into teams that can deliver continuous.”

Kenny Baas is a Software Engineer and Consultant focusing on software quality at Xebia Labs. He mentors teams by using practices like Domain Driven Design, Behaviour Driven Development, Test Driven Development and Continuous Delivery. He is an advocate for collaborating in open spaces. By using techniques such as Example Mapping and Event Storming, to engineer requirements and design of software along with business- and stakeholders. With these methods, he aims to create a transparent collaborative domain space with constant and instant feedback of the software that is being built.

Read more about Kenny
Jafar Soltani

Jafar Soltani

Adopting Continuous Delivery in AAA Video Games (Audience: INTERMEDIATE)


My talk is about how Rare changed its culture and approach for developing ‘A Sea Of Thieves’ by adopting Continuous Delivery.

As the result of this change, we’ve minimized crunch and have been sustainably delivering new features to players for the past 2 years. We’ve had over 120 releases so far, which means we can get feedback from players, evolve and improve the game. Our bug count has been very low (less than 50 high priority bugs) throughout the project. As we have invested heavily in automated tests, we’ve had a very small manual test team which has reduced cost significantly.

Adopting Continuous Delivery practices isn’t trivial and whilst we are several years into the journey, we are continuously working to improve our capabilities. I’ll discuss the challenges we’ve faced and the principles behind the techniques and practices we’ve used, such as:

– Changing culture so that everyone including Developers, Artists, Production and Test teams are responsible for the quality of the game.
– Changing culture to adopt extensive automated testing.
– Working in small batches.
– Adopting Continuous Integration practices such as Trunk based development, Feature Toggles and locking the depot when the Commit stage fails.
– Continuously improving the Commit stage to ensure we get fast and reliable feedback. I will cover the challenges in building AAA games in C++, transferring large game packages between agents, running 40,000 automated tests, identifying the most relevant and valuable tests to run as part of the Commit stage and dealing with flaky tests.
– Ensuring the game is always shippable. I’ll discuss prioritising fixing bugs and broken tests over developing new features. I will describe how we release regularly and how we’ve been releasing at least once a week over the last two years.
– Applying a continuous improvement culture, where we’ve been collecting telemetry from players and how we’ve used the data to help us better design the game.

Jafar joined Rare in January 2008. During his time at Rare, Jafar has worked on Banjo-Kazooie: Nuts & Bolts, Viva Piñata: Trouble in Paradise, Kinect Sports Series, Rare Replay. He’s most recently been leading the deployment pipeline team working on Sea of Thieves helping Rare adopting Continuous Delivery.

Read more about Jafar
Mark Hender

Mark Hender

What if you treat your CI pipeline as a product? (Audience: ALL)


There are many tools that facilitate a “CI as Code” approach to continuous delivery, but how many organisations are treating their CI infrastructure as an internally-facing product? Our software team has taken “CI as Code” quite literally over the last two years, as a dedicated team has developed a highly-configurable template for a Concourse pipeline. This template features opinionated APIs and useful abstractions that enable product teams to build, test, and ship open-sourced releases as well as artefacts for enterprise platform users. This talk will be equal parts experience report and cautionary tale: We’ll share the technical and cultural challenges of shared infrastructure templates from the perspectives of both producer and consumer, and we’ll make some recommendations about when and how you should consider implementing this pattern in your organisation. When your pipelines are code, it needs love just like everything else you ship!

Mark joined Pivotal Cloud Foundry as a software engineer in September 2016. At Pivotal, he anchored a project to provide a generic best practice CI Pipeline-as-a-product to support teams who deliver data services to open source Cloud Foundry users and enterprise platform customers. Currently he is working on RabbitMQ for PCF. He speaks fluent YAML and Concourse.

Read more about Mark
Michiel Rook

Michiel Rook

Leadership in Agile/DevOps environments (Audience: INTERMEDIATE)


The Agile and DevOps movements place a lot of emphasis on the autonomy, self-organization and responsibility of teams. A common misconception is that there is little or no role for leaders of such teams.

On the contrary! In this rapidly changing world, where competition is fierce and the pressure to deliver is high, effective leaders create environments that allow teams to thrive. They enable, coach and inspire.

But what makes a leader an effective leader? Attend this talk to learn more about various leadership theories, how they can be applied in practice, and my thoughts on leadership based on experience as a tech lead and scrum master and from military service.”

Michiel Rook is an experienced freelance consultant, developer, trainer & speaker from the Netherlands. He loves helping teams and companies to develop better software and improve the delivery process. When he’s not thinking about continuous deployment, devops or event sourcing he enjoys music, cars, sports and movies.

Read more about Michiel
Pan Thomakos

Pan Thomakos

Rapid Mobile App Delivery (Audience: ALL)


If I asked you to envision a continuous delivery pipeline, you would probably first imagine a CI service, maybe a few build boxes, and then some cool release service that gradually balances traffic to the new version of your API or website. But what about your mobile applications? Don’t they also deserve the benefits of a continuous delivery pipeline?

We started out, as most startups do, with a fairly ad-hoc mobile release process. Eventually we settled on a rather manual process that took two weeks to complete. Android and iPhone releases were carefully shepherded through to completion, available in 14 languages, and led by two release conductors – and a multi-page instruction manual. But this past year we decided to blow things up. We decided to automate as much of the process as we could, and to cut our release time in half. Spoiler: it worked!

If continuous delivery for mobile apps has never really occurred to you, or if continuously releasing a mobile application simply seems like an insurmountable task, then come learn about how we achieved this within a year. A rapid mobile app delivery pipeline has opened up all the benefits that we typically associate with continuous delivery (quick A/B test turnaround, putting features in front of our users sooner) to our mobile applications. In this talk I’ll share our journey with you, so that you can bring the benefits of continuous delivery to your own mobile release process.”

Pan Thomakos leads productivity engineering at Strava. He spends most of his time developing automation to eliminate manual tasks, and improving engineering processes. He is originally from Greece and spends his free time running, cooking, and playing with his kids.

Read more about Pan
Robin Weston

Robin Weston

How to run a successful Continuous Delivery Enablement team (Audience: INTERMEDIATE)


Although a standalone Continuous Delivery team can be seen as an anti-pattern, sometimes one might be a necessary evil in order to bootstrap an organisation’s Continuous Delivery capability. In this session we’ll look at the lessons we learned from standing up and running a standalone Continuous Delivery enablement team within a large enterprise. How did we measure progress, use data to influence and change culture and process, spread skills and knowledge within the organisation, and ultimately render ourselves redundant?

Robin has worked in technology for over 10 years and has been at ThoughtWorks since 2014. He cares passionately about helping deliver value to clients through simple, effective software. Away from work he spends his time running long distances for no particular reason, planning murder mystery treasure hunts and continually pinching himself to check that Leicester City really did win the Premier League.

Read more about Robin
Sabrina Leandro

Sabrina Leandro

How to rewrite, a bit at a time (Audience: INTERMEDIATE)


There’s nothing more frustrating than not being able to deliver new features because of unnecessarily complicated code. A fast deployment pipeline is of no use if the code itself is too hard and slow to change.

You may decide it’s time to throw it all away and start over… and what began full of optimism, drags on for months and years, adding even more complexity and various levels of legacy.

It doesn’t have to be this way! In this presentation, you’ll hear a first-hand experience of how to approach technical debt in an iterative way. This talk will teach you lessons on how to start, and most importantly, finish a big rewrite project.

You will learn how to approach the conversation with the “business”, avoid the most common pitfalls when changing the architecture of a complex codebase, and ensure the rewrite brings value from the start till the very happy end.”

Sabrina moved to London from Brazil, and after spending many years helping build Songkick, a live music platform for fans and artists, she’s now working with companies to improve their tech and processes. She enjoys working in cross-functional teams developing useful and delightful products, while focusing on code quality and good software design practices.

Read more about Sabrina
Tommy Tynjä

Tommy Tynjä

A year of Mob Programming - tips and tricks


Mob programming is when the whole development team works on solving the same problem, at the same time, in the same space and with one shared computer, screen and keyboard. I’ve been working in a development team where we’ve been mob programming for well over a year, every day, without exceptions. With this approach, together with our Continuous Deployment setup, we’ve noticed an enormous boost in productivity and we really feel that we make the most possible use of the team’s overall brain capacity to solve problems and to ship high quality software to our customers. But it’s not always that easy. We’ve learned to master this way of working the hard way, by continuously improving on our processes. In this presentation I will give an introduction to mob programming, the positive effects we’ve seen, why you should consider this way of working and some tips and tricks that we’ve learned along the way.

Tommy is a Senior Software Engineer and Continuous Delivery Consultant at Diabol. An advocate of open source software, Continuous Delivery, test-driven development, automation and tools that boost developer productivity and get software shipped faster. Specializing in systems running on the Java platform. Has been contributing to various open source projects over the years, such as the integration testing framework Arquillian and Apache TomEE. Currently maintainer of the Jenkins Delivery Pipeline plugin. Co-founder of the Continuous Delivery Stockholm meetup group.

Read more about Tommy