The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
🍳 Darakai Live Kitchen — Reimagining Food Delivery
What if customers didn't have to simply wait for their food?
I designed Darakai Live Kitchen, a new concept for the Darakai Food Delivery app that lets customers watch their food being prepared live by the chef.
Instead of a simple “Preparing your order” status, customers can:
🔴 Watch the kitchen live
💬 Interact with the chef
❤️ Send reactions
🍽️ Follow the cooking → plating → ready journey
🚴 Track the transition from kitchen to delivery
On the restaurant side, chefs get a dedicated Kitchen Dashboard to manage orders, start a live cooking session, communicate with customers, and end the stream when the order is ready.
The experience
Order → Confirmed → Cooking Live → Plating → Ready → Pickup → Delivered
The idea is to transform food delivery from a simple transaction into a more transparent, engaging, and human experience.
Technical direction
The concept is designed around a Flutter customer app + restaurant interface + .NET API, with real-time video handled by dedicated streaming infrastructure while the backend manages orders, live sessions, permissions, chat, notifications, and session status.
More than delivery. A closer connection.
#Flutter #DotNet #FoodDelivery #UXDesign #ProductDesign #LiveStreaming #MobileApp #Darakai
An interesting way to make the waiting stage part of the product experience. Dinner gets its own behind-the-scenes feature.
Exactly Mykola! That was the idea behind it. Instead of treating the waiting time as a passive part of food delivery, I wanted to turn it into part of the experience—giving customers more transparency and a real connection with the restaurant and chef.
There’s also a lot of...
If the kitchen stream drops, will the order timeline stay visible? Customers could still follow their order while the video is unavailable.
Absolutely — that’s exactly how I’d design it.
The order timeline would remain fully available even if the live stream temporarily drops. The video session is treated as a separate real-time layer, while the order status is controlled by the backend.
So the customer could still...
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Challenges
View all
Related posts
Designed a dashboard concept for tracking net worth across every asset type — stocks, crypto, property, pensions, funds, cash, and liabilities — all in one place.
Instead of the usual pie chart or bar graph, the centerpiece is a dial: a single, glanceable gauge showing where your wealth actually sits. Behind it, a quiet grid of every asset category fades in and out of visibility, reinforcing that everything's accounted for even when it's not the focus.
Dark UI, a single confident green accent, and a lot of negative space — built to make a genuinely overwhelming task (seeing your entire financial picture at once) feel calm instead of cluttered.
wow the animation feels super smooth, what is this animation for?
Working on the homepage for Still Blooming, a wedding and floral studio, I explored two different layout directions for the hero section. Same brand, same copy, just a different structure and visual rhythm.
Option 1 : Left Aligned. Text sits on the left with supporting content beside it, giving an editorial, asymmetrical feel.
Option 2 : Centered. Everything is stacked and centered, giving a clean, balanced, more classic hero layout.
Would love your honest UX/UI take which direction feels more elegant, functional, and true to a wedding/floral brand, and why?
22 voted
41%
32 voted
59%
54 votes
Closed
Great Design :)
Icon cards for me, bouquets are such a visual purchase that a scannable icon beats reading "small/medium/large" every time. Did the compact chips version convert worse in testing, or are you still gathering data on that?