Abandon the idea that you are ever going to finish. Lose track of the 400 pages and write just one page for each day, it helps. Then when it gets finished, you are always surprised.

Forget your generalized audience. In the first place, the nameless, faceless audience will scare you to death and in the second place, unlike the theater, it doesn’t exist. In writing, your audience is one single reader. I have found that sometimes it helps to pick out one person—a real person you know, or an imagined person and write to that one.

If a scene or a section gets the better of you and you still think you want it—bypass it and go on. When you have finished the whole you can come back to it and then you may find that the reason it gave trouble is because it didn’t belong there.

If there is a magic in story writing, and I am convinced there is, no one has ever been able to reduce it to a recipe that can be passed from one person to another. The formula seems to lie solely in the aching urge of the writer to convey something he feels important to the reader.

a bad story is only an ineffective story.

Guadagnino brought them Challengers, which will be released this month. Reznor said, “He started us down a path, saying, ‘What if it was very loud techno music through the whole film?’ ” (This is exactly what it turned out to be.)“I wish I had his notes,” Ross said of Guadagnino. “His notes were so fucking funny on what each piece was meant to do.”“Oh, yeah,” Reznor said. “ ‘Unending homoerotic desire.’ It was all a variation on those three words.”

the tooling we choose has a profound impact on the work we do, and, in the best case scenario, becomes a standard for how we build products. This is why we put so much care into even the tiniest details in Linear.

Even when doing concept work, you often need to focus your efforts. The design concept should feel like an exciting evolution of the product.

I didn't adhere to a specific method during the exploration phase, but typically, each day I designed a complete set of screens and flows. One day might be dedicated to designing the Inbox view, while the next day I could focus on the roadmap and projects. Other days, I explored upcoming product features. During this process, I experimented with different iterations of the sidebar, visual styles, and colors, and then linked the screens together as a prototype to assess their functionality.

Through this process, I generated hundreds of screens and was able to narrow down a few major directions that resonated most. Around this time, I began sharing the screens with other designers and people within the company to gather feedback and additional insights.

Ultimately, we settled on the main design direction, and I created a few views to showcase it

We started with the concept design Karri had originally imagined, but it wasn’t fully figured out and needed some additional design work. We didn’t know how we would bridge the previous UI design with the new style or if the new design could support all of our application states and options. We were able to make some changes off the bat, such as updating the color system, while other changes had to be punted to later on, such as the different headers you come across while navigating the app.

It’s easy for the scope of UI redesign projects to blow up. Before we got too far down any one path, we needed to get some confidence on the right option to keep everyone focused. So we ran some stress tests (or crash tests if you want to be dramatic) before going into implementation and iterating with engineers. We tested three main focus areas: the environment, the appearance, and the hierarchy.

Our app runs on Electron, so our navigation needed to work not just on macOS and Windows as a native app but also in any browser. That meant that previous/next navigation buttons, history, and tabs needed to be easily removable to work with browsers. We tested a lot of options, from very condensed to more spacious configurations. I often relied on Apple standards, which also helped get close to the feeling of a native app.

I also spent time aligning labels, icons, and buttons, both vertically and horizontally in the sidebar and tabs. It was definitely a challenge given the amount of UI elements we have on this tiny surface. This part of the redesign isn’t something you’ll immediately see but rather something that you’ll feel after a few minutes of using the app.

Karri mostly worked with opacities of black and white during his explorations, which really helped him get results quickly and helped me understand the relationship he had in mind between the elements and their respective elevation and hierarchy. As our system relied on a set of variables, I worked with Andreas on our software engineering team to polish and iterate on both the core variables and the operations we apply to them to generate our aliases for surfaces, texts, icons, and controls.

A while back, we rebuilt the system for generating custom themes in Linear, using the LCH color space instead of HSL. LCH has the benefit that it’s perpetually uniform, meaning a red and a yellow color with lightness 50 will appear roughly equally light to the human eye. This makes it possible to generate more consistently good-looking themes, regardless of which base colors are used.

Yes, the theme generation system also supports a contrast variable which defines how contrasty a theme should be. This allows us to automatically include super high-contrast themes for users who need it for accessibility reasons.

Linear relies on a set of structured layouts that support the navigation elements and content. It integrates additional headers to store filters and display options, side panels to display meta properties, as well as the actual display: list, board, timeline, split, and fullscreen.When I joined the project, Karri had already gathered most of the app's views and their respective states, so I was able to run all of my tests quite effectively. I mostly worked by type of view (list, board, split, etc.) as I found it easier to focus and ensure that every decision worked in all cases.

We divided the project into five milestones:Stress tests: Following the series of explorations made in November 2023, we tested if the direction felt right in the main views of Linear: Inbox, Triage, My Issues, Issues List, Project, Cycles, Roadmap, Search.Behavior definitions: As the direction was refined, we documented and defined the behaviors of the main components of the app: sidebar, tabs, app headers, and view headers.Sidebar and chrome refresh: We implemented the first bits of the refresh on the sidebar, tabs, and view headers. We also improved the appearance and contrast of our theme for light and dark modes. We used a feature flag to allow for internal testing at this stage.Private beta: We started rolling out the new design in Private beta to get initial feedback. Once we felt comfortable, we began rolling out the changes to a percentage of workspaces each day.GA: We released the new UI to all workspaces.

We knew that in order to move quickly and ship our work successfully, we needed to dedicate time and team resources to it. We couldn’t treat it as a side project.

Each afternoon, we divided the coding portions into groups of two engineers while designers iterated on other parts of the project, building a pipeline for us to work from. This daily back-and-forth between designers and engineers helped us get the first working version of the new UI by the end of the week

Next, we worked on the Inbox. We redesigned notifications to be more centered around the notification type and emphasized the faces of your teammates. We simplified headers and filters to improve the overall navigation. We also reviewed comments alignments and harmonized the look of our buttons with the new themes.

We started using Inter Display to add more expression to our headings while maintaining their readability and kept using regular Inter for the rest of the text elements.

This paper maps concepts from AI alignment onto a basic, three step interaction cycle, yielding a corresponding set of alignment objectives: 1) specification alignment: ensuring the user can efficiently and reliably communicate objectives to the AI, 2) process alignment: providing the ability to verify and optionally control the AI's execution process, and 3) evaluation support: ensuring the user can verify and understand the AI's output.

the notion of a Process Gulf, which highlights how differences between human and AI processes can lead to challenges in AI control.

While there were undoubtedly people who came in for the food, worked 3 hours a day, and enjoyed their early retirements, all the people I met were earnest, hard-working, and wanted to do great work. What beat them down were the gauntlet of reviews, the frequent re-orgs, the institutional scar tissue from past failures, and the complexity of doing even simple things on the world stage. Startups can afford to ignore many concerns, Googlers rarely can. What also got in the way were the people themselves - all the smart people who could argue against anything but not for something, all the leaders who lacked the courage to speak the uncomfortable truth, and all the people that were hired without a clear project to work on, but must still be retained through promotion-worthy made-up work.

Another blocker to progress that I saw up close was the imbalance of a top heavy team. A team with multiple successful co-founders and 10-20 year Google veterans might sound like a recipe for great things, but it’s also a recipe for gridlock. This structure might work if there are multiple areas to explore, clear goals, and strong autonomy to pursue those paths.

Good teams regularly pay down debt by cleaning things up on quieter days. Just as real is process debt. A review added because of a launch gone wrong. A new legal check to guard against possible litigation. A section added to a document template. Layers accumulate over the years until you end up unable to release a new feature for months after it's ready because it's stuck between reviews, with an unclear path out.

To find jobs more quickly, students tend to showcase the types of work most in demand in our industry, making website and app redesigns the overwhelming majority of case studies being published.

The myopic focus on shorter time-to-job metrics requires schools to standardize their design curricula — a trend which is reflected in the structure of case studies being produced.

The UX bootcamp also benefits from its students’ projects being published on Medium: case studies become content marketing that helps attract new students — and therefore revenue.

How can you differentiate yourself when applying to a position, if case studies from other candidates look exactly the same at first glance?

formulas create the illusion that real projects will always follow those same predetermined steps. In reality, working with both in-house and agency teams is too chaotic to apply a one-size-fits-all methodology

Research budgets are not always available; user journeys are not always necessary.

Many case studies start with an unexplained “user need” that one infers is the student’s personal need (e.g. "an easier way to share music on Spotify"). User research is then utilized in an attempt to prove that other people experience the same problem.

Because students are used to ticking boxes on a standard design process template provided by their school, they forget to explain why they are utilizing a particular method. If everyone’s process is somewhat the same, what is your unique angle to the way you work?

Insights gathered from one step are rarely applied to the next one. As a result, most case studies feel complete, but few feel smart.

Many case studies describe the project process in great detail, only to conclude with a solution which is predictable, unpolished, and/or lacking insight.

There are only so many audiences a single case study can speak to. If the student’s main objective in creating a case study is to get a foot in the industry, it might be a good idea for them to prioritize recruiters and hiring managers when thinking about their public-facing output. Ultimately, recruiters are the ones who screen candidates and hiring managers are the ones who make the final decision.

Because their time is limited, most hiring managers do not go through a designer’s entire portfolio, but instead review a couple of case studies

Since they must review a high volume of case studies as part of the recruiting process, hiring managers rarely read thoroughly, but instead quickly scan students’ work in search of talent

First [I do] just a quick scan. If the work looks interesting then I proceed to a more thorough analysis. This can range from 30 seconds to 15 or 20 minutes, it all depends on the quality and quantity of the work. Weak portfolios get discarded in seconds. Really good portfolios are analyzed in-depth and I usually end up on the candidate’s website or Twitter account. — Head of Design, 13 years of experience

Another common concern raised by the upper-level managers we spoke with was the extreme focus on the part of students on the design process, at the expense of the quality of the output. They reported that this was especially the case with UX and product-focused portfolios.

I jump to the final designs, I want to see the outcome of the case study. Only then, if the final designs are solid I go back and try to understand the designer’s process. You see a lot of people building really deep use-cases, but the execution fails and right now I need designers who can think and deliver, not one or another. — VP of Product Design, 17 years of experience

Your case study structure should reflect the areas that most interest you as well as the ones that will help you reach your personal goals. What type of company do you want to work with, and what kind of story will make the design leadership of that organization excited about the possibility of working with you?

If UI design is one of your strengths, showing personas that are shallow or unresearched can hurt more than they can help.

If you are writing about a real project for a real client, it’s important to explain the constraints and limitations around which you were working. If there were no constraints involved —as in the case of a project completed as part of a UX course— hiring managers will expect the designs to be as as innovative as possible, and will be frustrated if they are not.

Focus on insights rather than process

Reading an insightful case study is much more exciting than reading one that is complete but uninspired.

Would someone feel compelled to retweet a random sentence from your case study?

how you brought those insights to each subsequent step of the process.

The case study should tell a story about how you think and how you design — it's never about the project itself. Ideally, your case study should reflect your personality so strongly that it would feel out of place in any other designer’s portfolio.

What excited you the most about working on this project when you first received the brief?

A case study should focus less on the project it portrays and more on the skills and personality of its designer.

it should provide a platform for you to talk about the things you believe as a designer: your passion (reflected in how much time and attention you invested in the project), your thinking (demonstrated by how you connect the dots throughout your process), and your insights as you learn new things and evolve in your career.