FILL — AI Cancellation Recovery for MIRROR 27
I built FILL for MIRROR 27, a boutique salon in Indiranagar, Bengaluru, to solve a simple but expensive problem: last-minute cancellations leave valuable appointment slots empty, and salon owners have to manually chase customers through WhatsApp to fill them.
FILL turns that entire follow-up process into an automated recovery flow.
How it works
A customer cancels an appointment.
FILL immediately understands the service, duration, stylist, time slot, and value of the opening.
It searches the salon’s existing customer/inquiry pool for people who are likely to take that exact slot.
Customers are ranked with an easy-to-understand match score.
The owner can review the best match and send an offer.
The customer receives a simple mobile experience: “A spot just opened for you ”
With one tap, they can claim this appointment.
Once accepted, the appointment is automatically confirmed, and the owner’s schedule is updated.
If the customer declines or the offer expires, FILL can continue with the next suitable customer.
The key idea
Don’t let cancellations become empty chairs.
Instead of giving salon owners another complicated dashboard or booking system, FILL focuses on the moment where money and time are actually lost — the follow-through after a cancellation.
The owner shouldn't have to search old conversations, remember who wanted a particular service, or send multiple WhatsApp messages.
FILL finds the opportunity, finds the customer, and helps turn the empty slot back into a confirmed appointment.
What I focused on
Simple owner experience
Minimal manual work
Customer-first mobile booking flow
Explainable AI matching
Shared Owner ↔ Customer appointment state
Cancellation, rescheduling and confirmation flows
Clear recovery feedback such as ₹3,200 recovered
Premium, approachable salon UI without making the product feel
overly technical
Business: MIRROR 27
Location: Indiranagar, Bengaluru
Product: FILL
Core problem: Empty appointment slots caused by last-minute cancellations
Solution: Automatically match and recover those slots with the
right customer.
2
1
177
A hospital workflow brings together patients, doctors, nurses and administrators. I built CareFlow AI in Lovable to explore how those experiences can work together.
The demo includes appointment booking, dedicated role-based workspaces and an operational dashboard, using fictional data.
Explore it here:
https://careflow-command-centre.lovable.app
I’m available for freelance projects involving workflow automation, dashboards and web app prototypes.
Have a process you’d like to simplify? Let’s connect.
#Lovable #WorkflowAutomation #WebDevelopment #OpenToWork
0
65
:Mushk is a fictional fragrance brand I originally created as a branding project. When I came across the Bubble Challenge, I saw it as an opportunity to bring the brand to life digitally. Using AI-powered tools and Bubble, I transformed Mushk's identity into a website that reflects its story, aesthetics, and user experience.🫧🛁
3
5
1.5K
Introducing Poster Grammar - an AI design engine built to move beyond AI slop.
Most AI poster tools generate variations. But generating more designs doesn’t necessarily mean creating better ones.
I built Poster Grammar to approach this differently. Instead of another template-based tool or a thin wrapper around an image model, it’s a design engine built to understand the fundamentals that make visual communication work:
-Composition & layout - creating visual hierarchy and balanced arrangements.
- Typography & fonts - making type an intentional part of the design.
-Marketing intent - understanding what a poster needs to communicate.
- Creative variation - exploring different visual directions while maintaining design coherence.
The goal is simple: move beyond generic AI-generated visuals and create posters that feel intentional, considered, and designed for a purpose. You can explore different creative directions, experiment with variations, and generate designs with a stronger understanding of visual language.
I built Poster Grammar because I believe AI design tools should do more than generate images. They should understand why a design works.
Try it out: https://postergrammar.com
0
11
Namma Kutumba Care ❤️
Even when family is far away, family stays with you.
When we were children, our parents and grandparents held our hands and taught us how to live.
But somewhere along the way, technology moved faster than they could keep up.
Missed calls. Confusing remotes. Forgotten medicines. Silent homes.
Not because family stopped caring — but because distance, work, and life quietly came in between.
That feeling inspired me to build Namma Kutumba Care — a retro-inspired care companion for senior citizens who struggle with technology and live independently.
Instead of forcing seniors to adapt to technology, what if technology adapted to them?
✨ How it works
📱 Family scans a QR code to remotely set up care preferences
👨👩👧 Configure family, doctors, medicines, reminders, emergency contacts & wellbeing support
⚡ Everything syncs instantly to the device in real time
Built with Figma Make + Supabase, enabling a connected care experience between family onboarding and the senior companion device.
Features include:
📞 One-tap audio/video family calls
❤️ Live wellbeing updates for caregivers
🚨 Emergency & safety support
👨⚕️ Quick doctor connectivity
🚕 Trusted rides for hospital visits
💊 Gentle medicine reminders
🎵 Familiar songs, bhajans & comfort routines
🌍 Multi-language support
For family, it’s not surveillance or analytics — but a warm daily story of care.
Built for @figma #ConfigMakeathon with one belief:
Technology should adapt to people — not the other way around.
🔗 Figma Make + Supabase Live Prototype — https://above-engine-55177178.figma.site
🎨 Figma Make Community Working File — https://www.figma.com/community/file/1648752487036095974/namma-kutumba-care-by-prakash-a-k-community
🎥 Story About Namma Kutumba Care Video — https://youtu.be/kpzydEioQpM
📱 Live Figma Make + Supabase Demo — Family Care Setup — https://youtu.be/favY8DguYWQ
Would love your thoughts ❤️
@Figma #ConfigMakeathon #AIForGood #SeniorCare #Accessibility #InclusiveDesign #DesignForIndia #HumanCenteredDesign #FigmaMake #Supabase
1
1
211
Reimagining Sports with AI — Introducing Fitzz
For the Figma Makeathon, I designed Fitzz, an AI-powered application for sports enthusiasts that brings sports discovery, community, performance tracking, athlete ratings, and personalized coaching into one unified experience.
Using Figma Make, I transformed an idea into a fully interactive product by breaking the design process into multiple phases—from ideation and user flows to AI-driven experiences and final visual refinement.
The goal was simple: help people discover sports, stay consistent, improve performance, and grow through AI-powered guidance.
Check it out on Figma Community, It's Live Guys!!
https://www.figma.com/community/file/1649133905560941364/fitzz-mobile-app-design
#FigmaMakeathon #FigmaMake #ProductDesign #UIDesign #UXDesign #AI #SportsTech #DesignChallenge
2
2
166
🚀 THE FUTURE FLASH
3 AI Wonders Making the Future Look WILD!
🤖 1. Meet Your New AI Teammate — Grok Bot!
Grok Bot is a new AI worker that can use a computer and log into tools to actually do jobs for you, instead of only chatting.
Mind-Blow Fact: 🤯 It can work like a digital teammate that gets tasks done while you focus on something else!
🕺 2. Robot Dance-Off! Unitree Is Going BIG
Unitree makes humanoid robots that can run, dance, and even perform martial-arts moves, and the company is now preparing for a huge stock-market debut.
Mind-Blow Fact: 🤯 Unitree's IPO was reportedly oversubscribed by more than 8,000 times!
🪄 3. Say “Build Me a Website!” — Figma Make Does It
Figma Make lets you describe a website in normal words, and AI can build the layout, interactions, and working code for you.
Mind-Blow Fact: 🤯 You can go from a simple idea like “make me a space-game website” to a working site without starting by writing code yourself!
🌟 THAT'S ALL FOR THIS EDITION!
The future isn't coming… it's already knocking on the door! 🚪🤖
1
1
82
Apollo Lead Enrichment Pipeline: Built a 13-node automation that pulls company and contact data from Apollo, enriches it with website scraping, scores leads using AI (Llama 3.3), and generates personalized outreach - all written back to Google Sheets automatically. Cuts hours of manual prospect research down to minutes.
1
122
Autonomous Employee Onboarding & Provisioning Agent
Technical Impact & Architecture:Intelligent Verification: An integrated LLM layer performs real-time semantic analysis on uploaded documents (I-9, W-4, NDAs) to verify signatures, legibility, and document type, achieving a 100% compliance rate.
Operational Scale: By automating the entire lifecycle—from secure folder creation to sequential "Welcome" communications—it reduces onboarding time from 10+ hours to less than 15 minutes per hire.
Strategic Value:
This agent doesn't just save time; it protects the company. By removing manual document handling, it eliminates human error in sensitive payroll and legal filings while providing a flawless, professional experience for every new employee.
1
3
168
AI-Powered Payments Platform — Operational Command Center for JP Morgan
About: JP Morgan's Payment Blotter was doing its job managing high-value institutional transactions across multiple teams but nobody could actually see what was happening. Layered permissions, fragmented listing logic, and role-based visibility rules that had accumulated over years meant the system provided data but not clarity. Operations associates, client service teams, and back-office stakeholders were all looking at the same platform and interpreting it differently. In a compliance-driven, time-sensitive environment, that ambiguity isn't a UX problem it's operational risk.
What I did: I was the sole designer on a 9-person cross-functional team, responsible for auditing the entire workflow architecture, untangling role-based logic inconsistencies, and redesigning the blotter into a structured operational workspace while simultaneously laying the AI foundation for future enhancements. 12 stakeholder interviews across 4 service locations in US and UK. 3 sprints. One mandate: make a dense enterprise system behave predictably without stripping the depth that power users depend on.
The key insight that drove every decision: in enterprise financial systems, clarity and consistency drive operational performance more than feature depth. The redesign surfaced role-aware access transparently, embedded ticket handling directly into contextual views, and simplified payment state logic so users could identify state faster, understand their access, act from the listing view, and share context without exporting.
Post Launch: Launched and awarded in 2025. Ticket actions down to under 3 minutes. Processing time under 4 minutes. Task satisfaction at 75–80%. Ticket volume per team brought under 2,000. A payment blotter that finally worked like a command center.
0
66
