If you have ever developed code in real software projects, you know it. Tests never run quickly enough, frequently enough, and reliably enough. I’ve been there myself — pull request validation tests that somehow manage to be both lengthy and useless. And the way it forces you to multitask!
In this webinar, I’m going to talk about how we’ve seen this problem manifest in different ways in different teams, and how Launchable enables you to run just the tests that matter, and shift tests left without hassle and ship quality code faster with less stress.
Over the years, I’ve worked on and off as a remote employee. Living in North Carolina the opportunities to work with tech startups are limited, but I’ve had the good fortune to work with several in the area and a few outside the area as a remote employee. At my latest gig, Launchable, we are fully remote with my Product Manager in Texas, both founders in California, and our development team in Japan.
I am a very collaborative person by nature. I really believe that great design is a product of multiple minds coming together to work on a problem. Ideas seldom spring from the mind fully formed. They require constant iteration and critique.
Great design is a product of multiple minds coming together to work on a problem.
Working remote presents its own challenges for collaboration. People can only handle so many hours of video chat. Meetings take an extra toll and are typically scheduled. This makes ad-hoc discussion difficult.
In the past, I’ve been the one pushing for more meetings with my remote team. It’s where I get feedback on ideas and build consensus around where we are heading.
But at Launchable, we have a big emphasis on a written culture. You could even call it non-verbal. It’s not that we don’t have meetings, it’s that we try to limit the number of repeating meetings each week. And often when we do have a meeting, it’s to review a document and finalize decisions that have been carefully articulated in written form well before the before the meeting begins.
One of the reasons we do this is that we are passionate about what we call deep work. Deep work is when you are in the zone, uninterrupted, and able to do your best work.
Deep work is when you are in the zone, uninterrupted, and able to do your best work.
Practically, one of the ways that we encourage deep work, is that we have two days a week that are designated for meetings (Tuesday and Thursday). The other three working days are reserved for deep work. We don’t hold to this policy strictly — we still schedule some meetings throughout the week (particularly when meeting with people outside the company) — but the policy encourages you think twice before scheduling an internal meeting on a deep work day. It’s also not uncommon to have several days a week with huge blocks of time open for more focused work.
What do we do on our deep work days? We are often writing or reviewing documents that other people have written.
There are a few other practices that help us here.
We use Confluence as our central knowledge repository and everything is recorded there. Our plans for the product, engineering specifications, insights on customer environments, marketing plans, customer handouts and documentation — you name it, it’s probably in Confluence.
We also insist that all major decisions have an associated Confluence page where the details are documented (we use the DACI format with pros & cons).
When people do have meetings we encourage folks to write notes and share them in Confluence so that those who are not at the meeting can still benefit from the discussion. This has the side benefit of making it less important to attend every meeting.
We also have a practice of recording daily check-ins on a long running Confluence page. You typically write a short paragraph on what you worked on that day. This becomes a kind of activity feed where you can learn what others are working on.
The last part of our Confluence strategy is that we encourage written conversations to take place in Confluence rather than via email. As documents are written and commented on, we mark threads as resolved that are no longer relevant. This avoids long and confusing email threads.
As a result of these practices, one change that is taking place in my thinking is the concept of urgency. Many times when I’m in the moment I find myself reaching for action. I quickly send some raw thoughts over Slack or ask to bounce something off a coworker.
But lately I’ve realized that doesn’t always lead to the best outcomes. Not every problem is urgent. Many problems are much easier to discuss after a little private thought and deliberation. Putting my thoughts down in writing first is also more respectful of my coworkers’ time.
Is there ever a place for the urgent? Of course. But does every question need an answer right way? Have I clearly thought through the issue before asking for input?
Does this mean that there is not a place for raw thinking? Do my thoughts need to be in perfect order before asking for input? Also no. Draft documents are a great way to ask for input. Writing helps focus the work and leads to more pointed feedback.
So what does this way of working feel like in practice?
At first, it definitely feels slower. In the moment it is. But it’s also much more intentional. It forces you to make your case in a thoughtful, well reasoned way.
It’s not a high-bandwidth rambling communication that wins the day, but rather precise well chosen words.
The funny thing about this form of written communication is that to be effective you need to be concise. People don’t have patience for anything else. It’s not a high-bandwidth rambling communication that wins the day, but rather precise well chosen words. Abraham Lincoln famously quipped that it took him much longer to write a five minute speech than an hour.
For me, the real benefit of working this way is that it leads to clearer thinking and better decisions. I’m learning that quality of thought is better than quantity of words.
We wanted Launchable to be a company that cares about work-life balance, doing more for the world around while having fun.
Rocket Fuel day was born to back our words with a structure that enabled that ethos. On this day, we collectively choose a theme every month and plan an activity for this day. The activity should help us step out from our daily work and nourish is in ways beyond work.
Last month, we visited “Color Me Mine” to paint ceramics in order to unleash our inner makers. We’ve been busy building a lot of “fuzzy” things like “healthcare for Launchable” so actually “making,” was a great reminder of the joy that is building stuff.
Color Me Mine is normally a place where children go, so the staff lady apparently wasn’t sure why two dudes are there on a Friday afternoon. It turns out she was half-Japanese and despite the fact that she was born & raised here, she still spoke a fairly fluent Japanese so naturally Kohsuke talked to her a bit.
Kohsuke went for a mug cup and painted it with a rocket. Harpreet went for a bigger two-piece dog painted to look like his dog – bonus points with his wife.
Eagerly waiting for the next rocket fuel day as it is around the corner. It happens every 3rd Friday of the month.
Lord Voldemort comes out of the shadows at Launchable
The inside news on Kohsuke starting Launchable
This is a repost of a Launchable internal blog to talk about Kohsuke starting Launchable. If you must ask, no – we don’t really think he is Lord Voldemort :-).
Finally, the d-day arrived where “you-know-who” or the “one who cannot be named” decided to show himself to the world. We have been tired calling him “you-know-who” in conversations in the last month or two.
Once, Lord Voldemort showed himself to the world, the entire wizarding world wrote platitudes about the fearsome magic he wields and the new spells he is about to invent at Launchable. The coverage in the wizarding magazines:
In preparation for this work, we mixed some potions to build a great marketing plan , creatively named “Kohsuke starting Launchable Announcement”. Our marketing contacts at our VC partners were impressed by the level of planning and coverage that this small team conjured up.
Part of the plan was getting our sorry looking website to the next level and doing our own blog on LaunchableInc .
There was a lot of firsts here for us a company
Media interviews: – Rosalie Chan@Business Insider – thanks to Arseny from Atlassian for a helpful connect; Alan Shimel@DevOps.com – our beloved Heidi Gilmore from CloudBees, John Waters@ ADTMag – hustle from Harpreet.
Twitter ads (if I manage to set that damn thing up),
podcast with Alan Shimel
Finally, a whole day blown watching media – we got paid to browse today.
Now that the Lord Voldemort is out of the shadows – what’s next?
Is there a showdown between Harry and Voldemort? (For those who don’t know, I go by Harry amongst friends).
Are they going to duel to death?
Is Launchable doomed?
Hold your horses, this isn’t the Harry Potter series.
In real life, Lord Voldemort and Harry are good friends.
Real life Voldemort is too much of a nice guy to carry the moniker around.
Thus, this is where the story of Voldemort ends. Kohsuke Kawaguchi emerges from the shadow to successfully tackle the Testing world.
The Voldemort doll is going to remain at the desk though 🙂
Software development, meet Big Data. That’s why Harpreet and I are starting Launchable.
As the creator of Jenkins, I had the unique privilege to go see many software development teams around the world. And a consistent theme has been emerging. Data is making profound impact to all kinds of businesses, where gut and instinct driven decisions are replaced by quantifiable metrics and models, which enables a whole new angle of process/efficiency improvements. Yet in the software development field, we still largely seem to be driven by our individual experiences, beliefs, and instincts. Quite simply, this IS the obstacle to a launch of your new software, hence the name “Launchable.”
Where this is really evident is testing. In so many projects, every pull request goes through a CI build that takes an hour or more. Then somebody comes in and does a code review, resulting in a minor touch-up that requires another whole hour of testing.
In other projects, so many tests are amassed and so many engineers are working on it such that it’s simply impractical to run every test against every change. So people rely on arbitrary subsetting of tests (think “smoke tests”) but those tests are chosen completely by instinct. And other tests are run on a fixed frequency (think “nightly e2e tests”).
I’ve been working on solving this problem by applying machine learning. At Launchable, we think we can predict with meaningful accuracy what tests are more likely to catch a regression given what has changed, and that translates to faster feedback to developers. What if you only need to wait for 5 minutes instead of one hour to get your colleague to review your code, because we can choose the right 10% of the tests that give you 80% confidence? What if instead of running 6hr nightly e2e tests that tests everything that changed in the past 24 hours, you run adaptive 2hr subset of e2e tests 3 times a day, and get feedback within the same day? If any of this sounds interesting to you, sign up to our beta invitation list!
As exciting as this is, there’s so much more that we can do to drive more efficiency in the software development process. We’ll talk more about that in the coming days. Follow us on Twitter, LinkedIn, Facebook, or our mailing list.
Another thing that excites me in this new chapter of mine is that I get to build a company and a business from scratch again. In 9 years that I spent with CloudBees, I learned a lot about how there’s so much more to building a successful startup beyond technology and engineering. This time around, as co-CEO, I intend to put that learning into practice. I’m on this journey with Harpreet Singh, who has been my partner in crime for what, maybe 15 years? Between the two of us, we have more than 25 years of combined experience in this space, we built a successful Jenkins business, and we learned to work together flawlessly. To me, working with people I respect and I enjoy counts a lot, and I believe a great team is a foundation that leads to a great outcome.
Finally, on that note, we are hiring. If you are passionate about software developer productivity, and interested in working with experienced entrepreneurs well known in the field who cares about the team, I think we have a unique value to offer. You know how to reach me.
We are a stealth mode startup, so this is probably the first time you’ve heard about us. In the last couple of months, we’ve been busy building our team, the company, and fund-raising. We are very excited to announce that we have closed our seed funding round and hit a key early milestone in our journey.
What do we do?
Development teams are paying millions of dollars running pointless tests. Besides the huge costs, these pointless tests don’t catch problems and consequently have a negative impact on productivity and morale of development teams. Launchable solves these infrastructure costs and productivity issues by identifying and running meaningful tests.
We are honored that our team and mission really excited top-tier Silicon Valley venture leaders, experienced entrepreneurs, and seasoned business executives who have supported us in this funding round.
We will reveal more of our plan as we go forward, so please subscribe to our mailing list, follow us on Twitter or LinkedIn so as not to miss these updates.
We believe that software teams are going to drive every significant human achievement. However, engineering teams continue to operate as craftsmen while marketing & sales counterparts have stolen the march on using data to drive massive gains.
Launchable’s mission is to transform these artisanal teams to data driven scientists delivering software confidently. We are bringing big data to the software development process and starting with the testing domain.
Increasingly, companies everywhere are using DevOps mantra to implement Digital Transformation initiatives. As DevOps thought leaders, we have seen companies go through multi-month/year efforts to improve their software delivery velocity.
In my past roles as Head of Product for CloudBees and Atlassian Bitbucket, I have had a front row seat to software teams struggling to improve their velocity. Hearing that they need to go through a 6-12 month cultural transformation is hard for these teams – they get it but the change isn’t happening fast enough. To top it, there is no guarantee that these changes will produce meaningful impact.
These DevOps transformation efforts are too slow and what really gets in the way of development teams are massive inefficiencies or waste in the testing process.
Waste manifests in forms like running tests that don’t matter or tests that are flaky. This waste burns through a team’s infrastructure budget and more importantly waste a developers time.
Launchable uses data and machine learning to eliminate these inefficiencies during testing and consequently improve a team’s development velocity.
About our investors and advisors
We have been very fortunate to have found venture partners and angels who have deep domain expertise and believe in our mission, our vision, and our team.
Dharmesh Thakker: Dharmesh has been the earliest cheer-leader for the team and encouraged us to start Launchable. As one of the leader investors in the DevOps space (JFrog), he has seen companies struggle to deliver software fast and has been waiting for teams to tackle the testing domain. He has invested in companies like Databricks, Cloudera, SumoLogic amongst others.
Max Schireson: Max was the CEO Mongo DB and now an EIR at Battery. Max has been a sounding board for our ideas. Max is energized that Launchable is helping address a pain point that will make developers productive.
John Vrionis: co-founder Unusual Ventures. John has worked with me from my past life at CloudBees. John deeply cares about developer productivity improvements and has often talked about the lack of innovation in the testing domain. I have been fortunate to be part of the Unusual Academy Gamma Cohort. John has invested in companies like AppDynamics, DataStacks, Mulesoft, Heptio, and CloudBees.
Sri Viswanath – CTO Atlassian. Sri brings in thought leadership in challenges faced by software teams at scale. Sri has been a leader in companies like Groupon, Ning and Sun Microsystems.
Noah Wasmer – Head of Technical Teams, Atlassian. As the head of industry defining products such as Jira, Bitbucket, OpsGenie, Jira Service Desk, Noah brings in a wealth of knowledge on the challenges faced software teams of all sizes. Before Atlassian, he was GM/SVP of the End User Computing Group at VMWare.
Sacha Labourey – CEO CloudBees. Sacha is a thought leader in the DevOps space and has helped organizations lead and scale DevOps transformations. Before CloudBees, Sacha was a co-founder of JBoss.
Masaru Tange – Founded and IPOed SHIFT, Inc. Tange-san brings a wealth of experience in the QA domain. SHIFT is a QA company and incidentally doubling revenue every year.
Koki Uchiyama – Founded and IPOed Hottolink, Inc. Uchiyama-San brings in experience on data and AI to Launchable. Hottolink uses data, AI to help teams in decision-making. He is well known for mentoring entrepreneurs.
Jagmeet Lamba Founder and CEO Certa. An early cheerleader of Launchable who deeply believes in the founding teams capability to innovate in the testing domain. Jagmeet brings experience helping navigate enterprise sales and ML.
Bob Bickel – CEO RunSignup, Advisor CloudBees, founder Redline13. Bob is a 30 year veteran in companies like BlueStone Software (HP). Bob has really done it all in the enterprise middleware, DevOps and testing space. That plus, he was the Apple Pie making champion in 1999.
Jay Manni – A cyber security veteran – VP of Eng for Dtex. He was the founding engineering leader in Fireeye and helped take it from the first few to IPO.
Michel Goossens – Angel Investor, Advisor and former Head of Sales CloudBees. Michel is the Guru in helping setup and scale enterprise sales teams and motion. He was also one of the early players in Magento.
Steve Harris – Advisor and former Head of Product CloudBees. Steve is an industry veteran and a product leader in the enterprise, middleware and DevOps space. In his past life, he was “Mr. Java” at Oracle and was the head of the Oracle WebLogic business unit.
Andrew Lee Strategic Investor and Acquiror at Samsung SDS. Andrew brings in experience in M&A, partnerships in the enterprise software business. In his past life , Andrew has led the SpringSource acquisition into VMWare.
I am very excited by this opportunity and the Launchable team to help customers deliver software confidently.