I looked around my room the other day and noticed something obvious that I had somehow never noticed before. Every surface is flat. The walls, the ceiling, the floor — all plane, all smooth, all designed for exactly one thing: to look clean. Nobody who built this room thought about what it sounds like to be inside it.
That seems like a small thing until you stand inside Gol Gumbaz.
Gol Gumbaz is a seventeenth-century mausoleum in Bijapur, a few hundred kilometres from where I live. Its dome is one of the largest in the world, and inside it sits a whispering gallery — a narrow walkway high up along the dome’s inner wall where one small whisper against the surface can be heard clearly on the opposite side, nearly forty metres away. This is not a trick. It is engineering. The dome’s curvature was designed not just to look a certain way but to move sound waves along a precise path. Form and acoustics were treated as the same problem, not two separate ones.
And Gol Gumbaz is not an outlier. Across South India, there are temple halls with stone pillars that produce distinct musical notes when struck — each pillar carved to resonate at a different pitch, mimicking a different instrument. Someone shaped solid rock to sing. Not because they had better technology than we do. Because they were paying attention to something we have since decided to ignore.
I think about this when I sit in a coworking space.
I work out of one of the better ones available in my city — world-class interior design, good furniture, the kind of place that photographs well for a website. And yet, acoustically, it is chaos. An open floor, hundreds of people talking over each other, sound bouncing off hard surfaces with nowhere to go. You can spend lakhs on lighting and layout and ergonomic chairs, and still build a room that feels exhausting to sit in for eight hours, because nobody designed for what the room sounds like. Only for what it looks like.
This is the part that keeps pulling at me. It is not just an architecture problem. It is a pattern in how we think about design itself — maybe even how we think about quality in general. We have become very good at optimising for the visible. The things you can see in a screenshot, put in a presentation, point to in a meeting. And we have become remarkably bad at caring for the invisible — the things that are felt, not seen. Sound is the clearest example, but it is not the only one.
I see the same pattern in my own work. In data science, the dashboard is the visible output. It gets the attention, the feedback, the design reviews. But underneath it sits data quality, pipeline reliability, the assumptions baked into a model — none of which are visible to the person looking at the chart. And when those invisible layers are weak, the dashboard still looks fine. It just quietly means nothing. The numbers are there. They are even correct, technically. But the foundation they rest on was never given the same care as the surface.
I think this happens because we have trained ourselves to evaluate things by how they appear, not by how they perform in the dimensions we cannot immediately perceive. A beautifully designed room that sounds terrible is still called a beautifully designed room. A clean dashboard built on messy data is still called a good dashboard. The invisible failure doesn’t register — not because it doesn’t matter, but because we have lost the habit of checking for it.
The builders of Gol Gumbaz did not have that blind spot. They understood that a dome is not just a shape — it is a space that people will stand inside, and that standing inside it will be an experience made of light, proportion, and sound. They designed for the full experience, including the parts that don’t show up in a photograph. The temple builders who carved musical pillars understood something similar: that a hall is not just a structure, it is an environment, and an environment includes what you hear.
Somewhere between the seventeenth century and now, we narrowed our definition of good design to mean good-looking design. And everything that falls outside the visual — acoustics, airflow, the way a space feels after you have been in it for hours — became an afterthought. Not because we lack the technology. We have more of it than those builders could have imagined. But because we stopped asking the question they started with: what will it be like to be inside this?
I don’t have a fix for this. I am not an architect or an acoustic engineer. But I do think the observation matters — that the things we neglect are almost always the things we cannot see, and that this neglect is not limited to buildings. It runs through software, through data, through most of the work we do. We design for the eye and forget the ear. We build for the screenshot and forget the experience.
Maybe the first step is just noticing the flat walls.