Exponent energy · Shipped 2024

Designing interface for seamless self-service EV charging for multilingual audience

Designing interface for seamless self-service EV charging for multilingual audience

How I visualized invisible physics visually to enable fully autonomous charging for a multilingual audience, scaling the network to 50+ stations.

How I visualized invisible physics visually to enable fully autonomous charging for a multilingual audience, scaling the network to 50+ stations.

Timeline

Oct 2024 - Dec 2024

Skills

Product Management
User Research

Interaction Design

System Thinking

Motion Design

Product Management
User Research

Interaction Design

System Thinking

Motion Design

Role & Team

Lead Product Designer and collaborated with 1 Data Science Lead & 2 Engineers.

Responsibility

As the lead designer, I spearheaded the critical transition from operator-assisted to self-service charging to unlock network scalability.
I partnered with industrial designers to replace text-heavy instructions with a universal 3D visual language. This redesign addressed hardware anxiety, solved for multilingual accessibility, and implemented error-handling loops that directly enabled the removal of human operators from the loop.

Company

Exponent is India’s fastest EV‑charging startup, building a full‑stack battery and charging platform that can charge EVs from 0–100% in 15 minutes. It builds both the battery pack and the charging infrastructure as a coordinated stack, allowing real‑time control of cell conditions.

Overview

Unlocking scale through self-service.

Unlocking scale through self-service.

Exponent’s rapid charging technology fills a battery in 15 minutes using an off-board cooling system. The V1 chargers were bulky and complex, requiring trained human operators at every station. To scale effectively, we needed to transition from "manned" to "unmanned" stations.

My goal: Redesign the digital interface to make this high-voltage process safe, intuitive, and entirely self-service for first-time drivers.

Off-board cooling system

Solution

Solution

A text-independent, visual-first interface that visualizes invisible cooling processes.

We moved away from text-heavy instructions to a universal 3D visual language. By visualizing the invisible "off-board cooling" process through motion, the new interface gives users immediate confidence in the machine's safety without requiring them to read a single word.

The Main connection guide loop

Charging initiation showing the water flow

Charging showing water and electricity inflow

Unlocking and removing connector

Outcomes

Outcomes

50+ stations scaled with the capacity to manage unmanned operations, even in high-traffic locations, enhancing our ability to serve more regions efficiently.

Impact

38%

stations unmanned in 3 months

35%

reduction in charging errors

95%

driver satisfaction

The Challenge

The "Mess" of V1 Hardware

The "Mess" of V1 Hardware

Our first-generation chargers were beasts. They were heavy, utilized complex water-cooling, and the V1 interface was designed for trained employees, not everyday drivers.

To transition from "manned" to "unmanned" stations, we had to overcome two massive inherent constraints:

Physical Complexity

The connector is significantly heavier than a standard EV plug and requires a precise locking mechanism.

Off-Board Cooling

Unlike standard chargers, ours circulates water into the car, creating a complex cycle of "locking" and "draining" that most drivers had never encountered.

Research

Talking to the experts on the ground

Talking to the experts on the ground

I started by interviewing station operators and testing with 5 Exponent drivers to see how they handled this hardware. 





The results were clear: The current interface weren't helpful.

Current interface

Docked stage

Charging stage

Charged stage

Undock stage

Error stage

We uncovered three critical failures:

The "Dashboard vs. Pump" Conflict

When the car's dashboard hit "100%," drivers immediately assumed they were done. They would yank the cable ignoring the "Do Not Remove" text on our screen while the system was still draining hot water.

Language Barriers

In multilingual cities like Bangalore, text instructions were ignored. If the screen said "Align Connector," users just jammed it in randomly.

The "Invisible" Process

Because users couldn't see the water draining or the pins locking, they assumed the machine was broken or stuck.

Testing with the Drivers

Key Insight: We couldn't rely on text warnings. If the process is invisible (water draining), the user assumes it’s not happening. We had to make the invisible, visible.

Design Evolution

Iterations : "If they won't read, show them."

Iterations : "If they won't read, show them."

Our first breakthrough was realizing we needed to bridge the language gap. We decided to prototype using short, looping videos to demonstrate physical actions. I initially shot and edited short, looping videos of actual hands performing the steps (plugging in, turning keys).

First Iteration - Main Loop

Fun Fact: I was the hand model for these shoots!

First Iterations - Other Cases

Key On/Off Sequence Not Followed

Incomplete Connector Insertion

Solenoid locking issue

Charging Start Delay - Water entering

Charging

Charging Completion and Water Drainage

Premature Connector Removal

Handling Power Outages

We experimented with overlaying step numbers on these videos, but we realized that while video helped with instruction, it didn't solve the anxiety of invisible processes. While the videos were clear, they felt disconnected from the futuristic brand of Exponent. We needed something that felt native to the interface but retained the clarity of video.

The Pivot to 3D

The Pivot to 3D

While the real-life video helped with instruction (how to plug in), it didn't solve the anxiety of invisible processes (is it draining?). Real video couldn't show the water flowing inside the cable.

We decided to pivot to 3D fluid animations to visualize the internal state of the machine.

Designing for Hardware Constraints

Solving for the "Edge Cases"

Solving for the "Edge Cases"

The biggest design challenge wasn't the "Happy Path" it was the hardware constraints. The system required specific physical actions that felt unnatural to users.

Constraint 1: The Solenoid Lock

Problem: Due to the cable's weight, users often didn't push it deep enough for the locking pin to engage. The system would just error out.


Solution: We created a looping animation with a specific "depth line" visual, prompting users to push harder until the line disappeared.

Push Connector

Constraint 2: The Invisible Coolant

Problem: Users ripping the connector out during drainage damaged the pins.


Solution: Radical transparency. We visualized the water leaving the car. By showing an active "draining" animation, users understood they had to wait, even if the battery was at 100%.

Water draining

Core Final Flows

A 3D Language

A 3D Language

To make the status undeniable, we collaborated with Industrial Design to create a universal visual language.

The Whole Charging Process

Incomplete Connector Insertion

Impact

Business Impact

Business Impact

₹ XX Lakhs Saved Annually By eliminating the need for human operators, we significantly reduced operational burn, allowing capital to be reinvested in infrastructure.

50+ Stations Scaled The redesign successfully enabled the network to expand to high-traffic locations without staffing bottlenecks.

UX Metrics

UX Metrics

38% Self-Service

Stations converted to fully self-service within 3 months of launch.

35% Error Reduction

Drop in charging errors, specifically related to connector alignment.

95% Satisfaction

Drivers reported higher confidence and a smoother, anxiety-free process.

Reflection

What I learned

Design for the hardware, not just the screen. The screen doesn't exist in a vacuum. The physical weight of the cable and the locking mechanism were just as much a part of the UX as the pixels I was pushing.


Visuals > Text. In a high-anxiety, high-voltage environment, nobody reads. Universal animations saved us from localization nightmares and prevented dangerous errors.


Ship fast, test real. We didn't stay in Figma. We went to the stations, held the heavy cables, and watched real drivers fail. That context was the only reason we succeeded.