Valentin Cervellera He/Him — Twitch
French Front End Engineer at Twitch. Val is passionate about animation in the web, design systems, and user interfaces …
If I had to pick one thing that snowballed into getting into design, it was probably when I signed up for the PlayStation Forums in 2005. This was before Twitter, and message boards were where you could find your niche communities. Not sure what forums are like these days, but back then, you could modify your forum signature, which would show at the bottom of each of your posts. I noticed a lot of people on the forum had small rectangular designs as their signatures. I wanted one, so I went down a rabbit hole to figure out how to make my own.
That led me to discover a niche community of signature ("sig") designers who would post their creations, and people would comment and rate them. It feels worth saying, since it seems rare online today, but everyone was friendly and constructive with their comments. People would also share tutorials of how they made their sigs, and I quickly realized I needed to get ahold of Photoshop.
Once I started creating sigs, I then discovered various design forums, including one that had a board for selling web templates. People were selling templates for between $10 – $100. I was probably a freshman in high school around this time, working on my family's dairy farm, and once I learned I could make money designing things in Photoshop, I was all in from there.
I don't remember what the catalyst was, but I eventually started to teach myself HTML and CSS. This was the Internet Explorer 6 and table-based layout days. WordPress started popping, and I started to teach myself just enough PHP to add content management capabilities into the frontend code I was writing. I found my first few freelance clients, probably through help wanted posts on design forums. Most of the work was customizations to existing WordPress themes, which was great because it taught me what well-written code looked like and more advanced WordPress approaches.
At this point, I'm heading into college, studying a mix of graphic design and object-oriented programming. I had some financial aid, but was still going to have a decent amount of loans to pay off after graduation, so I spent a lot of my free time building up a business that sold WordPress themes that I designed and coded. That was a solid crash course in business management, marketing, and customer service. My college required students to do three internships in order to graduate, so each summer I also had an internship doing something related to design or web development, where I gained more evidence that this was an industry I wanted to get into.
It varies. I'm fortunate to work at a company with a generous remote work policy. I tend to split my weeks between the Nava NYC office and working from home. I find the commute and day-to-day interactions when you're out and about in the city to be draining, so those work-from-home days are pretty vital for me to recharge and stay mentally healthy.
A typical start to my day, when I head into the office, usually looks like me waking up at 8am and immediately playing music on my phone — I'm basically listening to music the rest of the day, unless I'm in a meeting. It's a half mile walk to the subway, which is relaxing when the weather is nice and miserable when it's not. Depending on my mood, on the subway ride I'll read articles I've saved to Instapaper, listen to an audiobook, or scroll through the news on the NYTimes iOS app.
Once at work, my day is driven by whatever my calendar tells me to do. My team has been using the Clockwise Google Calendar integration to automatically carve out focus time in our days, so I usually have at least one 2 hour chunk of reserved focus time in the day where I can go heads down and focus on whatever design or technical task I'm tackling. It's a good day when I have multiple blocks of focus times marked on my calendar. I have a recurring 12pm "Lunch" event on my calendar so no one books a meeting with me during that time. If I don't eat lunch, I become an unproductive miserable person...so I don't miss lunch. Beyond those things, there's meetings with government partners, 1:1's and pair coding with colleagues, and project planning meetings to keep everyone aligned. My team is distributed across the U.S., so we conduct these over a mix of Google Meet/Zoom/Skype, and are constantly communicating over Slack.
I try to wrap up work by 6pm and then hop back onto the subway. My brain is usually pretty fried at this point, so on my commute home I just listen to music, read the latest news, and catch up on Twitter, which I don't tend to look at throughout the workday. Once home, I enjoy cooking as a way to disconnect and decompress.
For me, usually what I need from inspiration-seeking are timely ideas. For that, I go to books (and occasionally blogs) written by people way smarter and more experienced than me. I try to read a range of books associated with whatever active challenges I'm working through to grow my awareness and understanding of the topic. When working on the HealthCare.gov design team, I read books like Trapped in America's Safety Net, The Form Book, and Automating Inequality, which informed my approach and framing of that work.
A recent guilty pleasure of mine is the Philips Hue ecosystem, specifically their smart light bulbs. They're dumb expensive, still rough around the edges, and totally unnecessary for anyone to install in their apartment, but they're also pretty fun to nerd out on if you're into automating/optimizing routine parts of your life. I've set up a Flux-like lighting system in my bathroom where the lights are bright and white in the morning, and then they automatically shift to a dimmer yellow after sunset.
One of the frustrating things about "smart" lights though is that you can't control them or achieve this level of automation if your "dumb" light switch is in the Off position. That means you either need to install a "smart" light switch or you need to keep the "dumb" light switch in the On postion and only control the lights using the app. I realize none of this sounds like "great design" yet, but I'm getting to my actual answer...
I did an unreasonable amount of research into smart light switches, searching for something temporary that I could safely install myself in my apartment, without rewiring things. Ultimately the product I landed on was the Lutron Aurora switch, which you install overtop of your existing "dumb" light switch with a few twists of a screw, locking it in the On position. You can then turn the lights on and off, and dim them, using the Aurora switch. I thought this was a nifty hack that enables new functionality while piggybacking on existing infrastructure. It made me think of the legacy government systems that we often have to integrate with in the work we do at Nava, like a system at the Veterans Affairs that currently powers appeals processing and which has been around since the 80s. It'd be nice to replace everything at once, but that's often not feasible. That means you have to get creative with your solution, figuring out how to deliver a seamless and effective experience to citizens by injecting your solution in between them and the legacy infrastructure that was there before.
I'm proud of the work I've had the opportunity to be a part of over the past several years at Nava. I worked on a talented team to redesign and build the complex eligibility application on HealthCare.gov. During the Affordable Care Act’s 2018 open enrollment, more than 7.5 million eligibility applications, encompassing over 11 million Americans seeking health coverage, were submitted on HealthCare.gov. One of those applications was my Dad's.
At a high level, the application determines whether a person’s family is eligible for an insurance plan, and whether they’re eligible for any cost savings to make their insurance more affordable. In addition, the application determines whether a person is eligible for Medicaid or the Children’s Health Insurance Program (CHIP), programs which provide free or low-cost health coverage to millions of Americans. The application therefore determines whether or not people, both at the individual and household level, are eligible for a variety of programs and services. To make it even more complex, Medicaid and CHIP policies vary state-to-state. A lot of our team's time was spent untangling these various policies, determining what information we needed to ask for, and how we could streamline those questions so that we weren't asking for redundant or unnecessary information.
I was proud of the foundational design patterns that our team came up with, building on a lot of great work by the UK Government Digital Service and U.S. Web Design System. We spent a lot of time working through the proper sequencing of questions, how to break a complex form into digestible chunks, and how to respectfully and clearly guide people through the process using help content.
Another big piece of the work was making sure that we were designing the application to be simple and streamlined for the majority of use cases, while being robust enough to handle even the most complex household scenarios. For extreme cases that seem unlikely, it’s often tempting to classify them as an edge case and sweep them under the rug. However, for public services, improperly handling an “edge case” could mean the difference between someone being deemed eligible for health insurance or not. It’s irresponsible not to consider all the different ways the service you are building could break down, and our team spent a lot of time thinking through those scenarios. Wherever possible, we attempted to put safeguards in place so applicants could intuitively navigate through the experience without surprises or mistakes. If they do make a mistake, as we all occasionally do, we made sure to provide opportunities to fix them and continue where they left off.
The design patterns we implemented on HealthCare.gov in 2017–2018 are now being carried over and reused by the Integrated Benefits team that I'm on now, where we're aiming to improve the social safety net in the country, with a focus on making programs like health insurance and food assistance easier for residents to access and states to maintain.
Nava is a Public Benefit Corporation, where we partner with U.S state and federal government agencies to plan, design, and build their digital services. The work we do falls at the intersection of what some groups call Civic Tech and GovTech. The programs we’ve worked to support so far, from Medicare to Veterans Affairs to HealthCare.gov to Medicaid, provide benefits to over a quarter of the population of the United States.
Illustration by Vidhya Nagarajan
I've only been working in civic tech for about three years now, so my perspective on the tactical challenges we face continues to shift as I work on different projects. The challenges you face varies pretty widely depending on the project, ranging from policy and compliance hurdles that sometimes run counter to a positive user experience, navigating stakeholder and partner relationships, gaining access to end users for research, aligning teams around a service vision, and avoiding burnout. The programs we work on are super complex. Untangling them is no joke. That's a challenge, but one I personally enjoy working on. Every day is a new civics lesson. I'd recommend this post from Cyd Harrell if you're curious to learn more about civic tech's history, and what we now need more of.
Take advantage of internship opportunities. I wish I had experienced working at a range of company sizes earlier in my career. Each has their pros and cons.
Don't make your job your only focus. Have hobbies and participate in communities outside of your job. It'll make turbulent times at work a little less stinging. Also go the fuck home.
Be critical of the work, not the colleague who did the work.
Read books. Preferably not just books about your profession. I recommend So You Want to Talk About Race, by Ijeoma Oluo if you're looking for somewhere to start.
Other advice that I often come back to:
"Leave three unimportant things unsaid each day." – Kim Scott
Be wary of the structureless group. "Contrary to what we would like to believe, there is no such thing as a structureless group...structurelessness becomes a way of masking power." — Jo Freeman
Register to vote if you're eligible. If you're looking for a new job, check out designgigsforgood.org.