Reflecting on My Design Taste in 2026

When I look back at the portfolios I’ve built over the years, I can see how much my design “taste” has changed.

This week, just for fun, I spun up all the portfolios I’ve built over the years and went through them one by one. They made me think, “Did I really build this?” The first portfolio I built back in 2023 is probably the best example of that.

Where I Started

The colors feel off to me, the spacing and overall layout feel weird, and the performance wasn’t great either. It was built with Flutter. Even the project cards feel awful to me now, and the images I chose for the portfolio don’t look good to me anymore.

But I’m sure the 2023 Sai Charan thought this was a very good-looking portfolio because that was my design standard at the time.

Back then, we didn’t have agentic development, so this was completely hand-coded. I also know I spent a lot more time building it because I wasn’t as good at coding as I am now (I know I’m still not as good as I want to be, but I’m definitely much better than I was back then).

Looking back, I think both coding and design were bottlenecks for me at that point. I was still learning how to build things, but I was also still figuring out what I actually liked from a design perspective.

Finding My Style

Then, in 2024, I came across a framework called Kobweb, so I built my second portfolio using it.

By this time, I had also been exposed to a lot more design. I was more active on Twitter, following designers and looking at portfolios, components, and websites that were much better than what I had been building before. I started developing preferences and getting a better idea of the kind of design I liked.

One of those things was the Zinc theme from shadcn/ui. I really liked the color scheme, and honestly, I still do. It is probably one of my favorite color schemes even today, and I have a feeling it might become my default color scheme for future projects too.

So when I built the Kobweb portfolio, I knew I wanted to use Zinc and designed the portfolio around it.

Looking at it now, the color scheme is still something I really like. That’s probably one of the few things I would keep almost exactly as it is. The rest of the design makes me notice a lot of things I would change today.

I don’t like the spacing. There is a lot of white space between each section, and in general, there is just too much empty space. At the same time, some of the content feels cramped. The green chips for each project feel very out of place on the website. The source code button is too small, and the text spacing feels wrong.

But despite all of that, this was still a huge improvement over the previous portfolio. I was starting to develop a better visual sense.

Learning to Notice More

Then, in 2025, I rebuilt my portfolio again, this time using Flutter. I honestly don’t know why I built it in Flutter, but it was fun at the time.

The color scheme was still the same Zinc theme, and the spacing had improved a little compared to the previous version. But I still didn’t like the amount of white space between the sections. There was also a lot of inconsistency throughout the design. The section widths were inconsistent, and the About Me section also feels bad to me now. It is difficult to read because of the length of the text, and the use of bold text throughout the paragraph makes it even harder to read.

That said, I still like some parts of the design compared to the previous two portfolios, like the project cards and the footer. It was definitely moving in the right direction, but there were still a lot of inconsistencies and plenty of room for improvement.

And once again, the performance wasn’t great.

By this point, I was starting to understand more about what I liked and what I didn’t. I just wasn’t very consistent at applying it yet.

Where I Am Now

In 2026, I rebuilt my portfolio using Astro.

I went with the same Zinc color scheme again.

This time, I paid much more attention to spacing, consistency, and performance. By now, I’ve seen a lot more portfolios and websites with great designs, and I think I’ve developed a much better sense of what I like and what I don’t.

I also have AI to help me now, so building things has become much easier. I can focus more on thinking about the design itself instead of spending as much time figuring out how to implement every little thing.

Previously, both coding and design were bottlenecks for me. Now, I feel like design is the bigger bottleneck.

So I spent a lot of time thinking about each component and all the small quality-of-life improvements that most people probably won’t even notice. One of the places where I spent a lot of time on these details was the sidebar.

In all my previous portfolios, I had a top bar. I don’t think that was necessarily a bad idea, but this time I wanted to try something different.

I really like the ChatGPT-style sidebar rail navigation, and since my portfolio has multiple sections that are all part of the same page, I thought a sidebar rail would work really well.

There are also some small details in the sidebar that I spent time on. It has a small white indicator that shows the current section, and it can also be dragged to move between sections.

When I hover over Projects, it expands and shows all the available projects. When I move my mouse down toward Recent Posts, the Projects section would normally collapse, causing the Recent Posts item to move up. This can be frustrating because the user is already moving toward Recent Posts to click it, but because the Projects section collapses, the item suddenly moves away. So I made the Projects section stay expanded while the user is moving down toward the next section. This makes the interaction feel much smoother.

The same sidebar also changes on mobile. It turns into a small pill that I can tap to select a section or press and hold to scrub through the menu and navigate between sections.

There are also a few small details in other parts of the UI. For example, when I hover over the profile image in the top bar, a popover appears showing the current weather and time in my area. The background of the profile image also changes based on the current weather. I also added a small buffer around the popover, so if the mouse briefly moves outside the pill but stays within that area, the popover doesn’t immediately disappear.

Then there are all the other small things: button animations, press interactions, icon animations, transitions, hover states, and a lot of other details that I kept tweaking.

None of these things are particularly revolutionary, and most people will probably never consciously notice them.

But I enjoy thinking about them. That’s something I’ve developed over the years.

Those are things I wasn’t really thinking about when I built my first portfolio.

As I’ve grown as a developer, my design standards have grown too. I’m getting better at understanding what makes a good UI and UX and, more importantly, why something feels good or bad.

I won’t say this is the best portfolio out there.

But according to my standards right now, this is the best-looking portfolio I’ve created.

At the same time, I know my standards will continue to change.

I’ll Probably Hate This One Too

There are plenty of portfolios I’ve seen where I am completely blown away by how good they look. There are designers whose work makes me stop and wonder how they came up with certain ideas. Seeing work like that reminds me that I still have a lot to learn.

But that’s also part of what makes this fun for me. There is always something better to learn from, something new to notice, and something I can experiment with in my own work.

So I will keep looking, learning, experimenting, and building.

I’ll probably keep redesigning my portfolio. I’ll probably keep experimenting with different technologies. I’ll probably continue obsessing over tiny details that most people will never notice.

And eventually, I’ll probably look back at today’s version and think that it could have been much better.

Honestly, I hope that happens.

Because that would mean my standards changed again.

And maybe that’s the best way to know that I’m still improving.