“Where am I supposed to put these?”
“Is it safe to leave them here?”
“Will I make it back in time?”
Call
WhatsApp
Designed and developed by:
Mahindra Prasad Tudu
A city-scale luggage storage and last-mile delivery prototype for India's transit travellers. Built end-to-end in a 3-hour AI-accelerated Design Sprint, grounded in classic Design Thinking discipline.
India has the largest train network on Earth, three of the world's twenty busiest airports, and zero infrastructure for the most basic transit job: putting your bags down. I had carried this observation in my notes app for over a year. In March 2026, I gave myself a single afternoon to prove the problem was real, the solution was buildable, and that design thinking translated to AI-accelerated execution.
Design Thinking starts with empathy. Even in a compressed sprint, this phase cannot be skipped — it can only be condensed.
It started at Bengaluru City Railway Station. A young couple, four suitcases, arguing with an auto driver who wanted ₹600 to take them 2 km because they had nowhere to drop their things. Their flight was the next morning. A hotel meant paying for a full night. The station cloak room was closed.
I sat with that observation for a year before I designed anything.
Thirty minutes of structured research with Claude as a research partner. Two findings did most of the work.
A vague problem produces a vague product. To move beyond observation, structured interviews with the user segments most affected:
This section will update with real findings as interviews complete.
Based on observation and secondary research — to be refined with interview data.
A vague problem produces a vague product. Before any screen, three artifacts: a problem statement, three POV statements, and five How Might We questions.
India's 27+ million daily transit travellers need a trustworthy, on-demand way to store and move their luggage because existing infrastructure is unreliable, restrictive, and dignity-stripping — and global solutions like Bounce, Stasher, and Radical Storage have not entered the Indian market.
Anaya, a business traveller, needs to drop her bags within 15 minutes of arrival because her conference starts before her hotel will check her in.
A backpacker arriving overnight in Bangalore needs to store gear without a same-day train ticket because she's planning to bus to Coorg the next morning.
A wedding family on a layover needs to store and later deliver six suitcases to a venue because dragging them through a city for 8 hours is not an option.
Five questions that drove every screen.
Synthesized from observation, secondary research, and conversations — not a single real interview subject. To be replaced by a research-grounded persona after the planned interviews.
Six-stage map of Anaya's day without GoBags — emotion curve dipping into frustration during Realize → Workaround.
Twenty minutes to diverge — to consider concepts I would later reject. Showing rejected directions is the difference between a case study and a sales pitch.
Four directions, evaluated against viability, desirability, and feasibility for a 3-hour build.
Partner with hotels to store luggage in their cloak rooms.
Why rejected: Same trust failures as today. Can't reach non-guests. Replicates existing infrastructure rather than replacing it.
Pickup at origin, ship to destination.
Why rejected: Requires national logistics, multi-day operation. Wrong job — solves shipping, not the transit-day gap.
Bounce-style network with verified storage points.
Why picked: Asset-light. Scales fast. Solves the discovery problem. Maps to a proven international playbook.
Standalone kiosks at stations and airports.
Why deferred: Capex-heavy, slow rollout. Considered for v2 after the marketplace proves demand.
No Figma in this project. Exploration happened in two media:
GoBags app
├── Onboarding (4 screens)
│ ├── Welcome
│ ├── Aadhaar consent
│ ├── Permissions
│ └── Map handoff
├── Discovery
│ ├── Map (default)
│ ├── List view
│ └── Filters
├── Storage detail
│ ├── Photos
│ ├── Hours · Capacity · Rating
│ └── Reserve
├── Booking
│ ├── Duration & price
│ ├── Payment (UPI default)
│ └── Confirmation + QR
└── Add-on
├── Delivery scheduling
└── Delivery trackingA prototype is a question made testable. 18 hi-fi screens in HTML and CSS with Claude AI as co-pilot. The discipline came from the previous four phases — the speed came from AI.
Most case studies show Sketch → Wireframe → Hi-fi. This project skipped the wireframe entirely.
The most modern build flow possible in 2026. No intermediate Figma. No wireframe round-trip. It works because the upstream phases produced enough clarity that prompts could be specific. Vague phases produce vague prompts. Discipline upstream is what makes execution this fast downstream.
Four decisions, each shown with the rejected alternative.
Why: Storage requires bilateral trust. Aadhaar is the only ID layer that scales in India and is familiar to the user. Also a competitive moat — global entrants would have to build this from scratch.
Why: Transit users think in destinations, not brand names. The map answers the right question instantly. Pattern borrowed from Uber/Ola; familiarity earns 4 seconds of attention.
Why: UPI is what 80%+ of Indian app users actually transact with. Showing five options creates fake choice and adds 6–8 seconds to booking. Optimize for the dominant rail.
Why: Storage is the immediate pain. Delivery is upsell. Selling the upsell first confuses the user, the value prop, and the unit economics.
10 of the 18 have been published. Once you map which screen belongs to which flow step, captions will sit alongside each.
A prototype that hasn't met a real user is still a hypothesis. Here's the validation plan.
Three things I already suspect will need work, even before testing:
The counts are honest. The honesty is the differentiator.
18 screens, two flows, and India-specific patterns in roughly three hours. Claude didn't just write code — it caught edge cases, suggested empty states, and rendered locale specifics. But it didn't choose the persona, frame the problem, or pick Aadhaar over phone-OTP. Strategy stayed mine.
Tens of millions of daily transit passengers. Zero scaled city-storage solution. The market Europe solved a decade ago is still wide open here.
The compressed sprint worked because of the design thinking phases, not despite them. Skipping Empathize would have produced a prettier wrong product.
This idea sat in my notes for a year. Three hours to ship made it real, shareable, and defensible. The gap between idea and prototype is now closer than the gap between prototype and real product.