Amplifon ai

Amplifon ai

Year

2025

Category

product design

What Is Volta

Project Overview

Volta was an early-stage fitness tech startup aiming to bring social connection back to physical activity. The vision was to let people easily find workout partners, organize matches, book courts, and rent equipment ,all in one seamless mobile experience.

My challenge was to turn a raw business idea into a coherent, user-centered MVP that people would actually use, not another dead fitness app.

I was responsible for the entire UX strategy and product design, from defining the user problems and journeys to building scalable, accessible components and testing usability across flows.

What Is Amplifon AI

Project Overview

Amplifon is the global leader in hearing-care solutions, operating thousands of clinics worldwide. Their hearing-aid journey spans clinical interviews, device fitting, trial weeks, adaptation, calibration, and long-term care.


Their problem was systemic:
the admin side was producing unreliable data, and the user side lacked emotional and empathic guidance.
Both failures created drop-offs, poor personalization, and unnecessary clinic visits.


I redesigned the entire ecosystem into a two-sided, AI-powered platform:

  • Admin Side: AI-guided interview + structured data system that eliminates human error and builds a reusable personalization backbone.

  • User Side: A voice-first companion that supports patients through adaptation, calibration, personalization, and daily emotional guidance.


The result is a unified, scalable system that improves data quality, reduces operational friction, and delivers a more supportive, human experience.

Core Challenges


Failure 1 - The Data Layer Is Broken (Admin)

Business & Clinical Constraints:

  • Medical decisions rely on intake data that is currently incomplete, inconsistent, and unstructured.

  • Clinicians face time constraints, liability pressure, and cognitive overload.

  • Regulatory requirements demand accuracy and auditability.

Result:
Poor tuning, repeated clinic visits, and personalization that collapses after short time.


Failure 2 - The Emotional Layer Is Missing (User)

Behavioral Reality:
Users delay hearing aids for ~7 years due to fear, stigma, and emotional denial.

During trial week and adaptation:

  • new sounds are overwhelming,

  • support is minimal,

  • app UX is too complex for eldery, young users,

  • elderly users feel abandoned.

Result:
Satisfaction drops after 1–2 months → abandonment → expensive support loops → low retention and usage rate.


Strategic Approach

The redesign solves both failures through a connected system:

A- Admin Platform:

AI-assisted data intake that produces structured, auditable, reusable information powering years of personalization.

B- User Companion App:

A voice-first, emotion-aware companion guiding users through adaptation, calibration, and daily progress.

C- Shared Data Foundation:

A cross-platform schema linking both sides, enabling dynamic personalization and reducing manual intervention.

D- Safety & Alignment:

AI is assistive, not authoritative:

  • All critical decisions require clinician approval.

  • Structured schema reduces hallucination risk.

  • Coverage engine prevents missing medical data.

Core Challenges


Failure 1 - The Data Layer Is Broken (Admin)

Business & Clinical Constraints:

  • Medical decisions rely on intake data that is currently incomplete, inconsistent, and unstructured.

  • Clinicians face time constraints, liability pressure, and cognitive overload.

  • Regulatory requirements demand accuracy and auditability.

Result:
Poor tuning, repeated clinic visits, and personalization that collapses after short time.


Failure 2 - The Emotional Layer Is Missing (User)

Behavioral Reality:
Users delay hearing aids for ~7 years due to fear, stigma, and emotional denial.

During trial week and adaptation:

  • new sounds are overwhelming,

  • support is minimal,

  • app UX is too complex for eldery, young users,

  • elderly users feel abandoned.

Result:
Satisfaction drops after 1–2 months → abandonment → expensive support loops → low retention and usage rate.


Strategic Approach

The redesign solves both failures through a connected system:

A- Admin Platform:

AI-assisted data intake that produces structured, auditable, reusable information powering years of personalization.

B- User Companion App:

A voice-first, emotion-aware companion guiding users through adaptation, calibration, and daily progress.

C- Shared Data Foundation:

A cross-platform schema linking both sides, enabling dynamic personalization and reducing manual intervention.

D- Safety & Alignment:

AI is assistive, not authoritative:

  • All critical decisions require clinician approval.

  • Structured schema reduces hallucination risk.

  • Coverage engine prevents missing medical data.

Hackathon Version

First Iteration

Ideated, created and prototyped through the ESCP hackathon with collaboration with Amplifon in a 5-hour window.

Mobile Image
Mobile Image

Admin Platfrom V2

I redesigned and recreated the whole user journey, in this version The admin platform takes the patient’s interview, organizes everything automatically, and turns it into a clear, structured profile. It helps clinicians catch missing information, avoid mistakes, and build a reliable foundation for accurate tuning and long-term personalization.

AI-Guided Interview System

Impact: Clean, reusable, standardized data that finally enables long-term personalization.

Coverage & Completeness Engine

Impact: Zero skipped questions → complete intake every time.

Structured Patient Profile

The personalization backbone instead of free-text chaos.

Linear-Inspired Admin UI

Faster workflows, lower cognitive load, consistent data across clinics.

Innitial Step

The specialist fills the form with mandatory data from the patient like name, contact, medical history, and…
this part is handled by human to prevent errors.

AI instead of Human Notes

The first touch point is the interview, in the traditional way human specialist took notes of the interview and import the text by typing and storing.

Now AI listens, transcribes the dialogue, keeping them ready for the next steps of the process.

AI-assisted interview

Meanwhile, during the interview process, the AI agent starts sorting the collected data into categories, which will be helpful for personalization, support, and calibration of the devices.

At the same time, it provides suggested questions to ensure we cover all the necessary aspects and information for a seamless, user-centered, empathic, and data-driven user experience.

More than an Interview tool

Implemented features for a seamless experience

HR, Financial, Apoointments and Device management

View and manage all different areas and dashboards

Customer Satisfaction Alerts

Recieve alerts when customer satisfaction is dropping and respond with personalized solutions, like in app messages, notifications, calls and schedualed visits

Appointments CRUD

take control over scheduling and managing appointments

More than an Interview tool

Implemented features for a seamless experience

HR, Financial, Apoointments and Device management

View and manage all different areas and dashboards

Customer Satisfaction Alerts

Recieve alerts when customer satisfaction is dropping and respond with personalized solutions, like in app messages, notifications, calls and schedualed visits

Appointments CRUD

take control over scheduling and managing appointments

User Platfrom V2

The user platform guides patients through their hearing-aid journey with a friendly, AI-driven empathic companion that explains what to do, checks in daily, and helps them adapt comfortably. It reduces confusion, supports them emotionally, giving personalized advices and messages, and makes calibration and progress easier without needing constant clinic visits.

AI Companion

Value: Reduced fear → higher compliance → lower abandonment.

Personalized Onboarding Plan

Value: Users understand “what to test” → trial week success increases.

Daily Comfort/Clarity/Fatigue Check-Ins

Value: Fewer unnecessary clinic visits → faster adaptation.

In-App Auto Calibration

Value: Operational efficiency + reduced workload for clinics.

Context-Aware Personalization

Daily moments feel supported → retention increases.

Background Image

Authentic Connections

users can build genuine relationships with the character

Empathetic Partner

It will be user's emphatic and emotional guide

Personalized Experience

Tailored content and recommendations for users.

Hi Megan, How are we doing today?

The onboarding journey can be fully personalized using interview insights and extracted data, tailored to factors like age, gender, lifestyle, and more. Users can create, name, and engage with an AI companion that checks in, converses, and provides the empathetic, emotional support often missing in standard app interactions.

Background Image

Smart Search

The AI companion is embedded directly into the core of the app, especially in features like smart search. Whether through voice or text, it can scan all relevant data, execute tasks on the user’s behalf, or guide them step-by-step to the right action.

For example: ask, “How can I reset my settings?” and it immediately leads you to the exact place or handles the action if possible.

Lets have a talk!

The feature moves from simple, empathetic conversations about the user’s struggles to tailored recommendations, daily micro-tasks, and guidance that speeds up their adaptation to the device. It uses behavioral signals and extracted data to deliver targeted suggestions, making the overall experience smoother, more supportive, and genuinely user-centered.

The Call

The feature activates automatically based on behavior signals, satisfaction levels, and retention patterns. Its job is to detect user struggles, surface issues early, and deliver tailored feedback that smooths the entire experience. It boosts motivation, reassures users they’re supported, and encourages ongoing interaction with the companion. It also helps parents guide their children’s device use, and gives less tech-confident elderly users a simpler, more effortless journey.

Prototyping

I leveraged Cursor, MCP servers, and Figma Make for rapid prototyping, enabling a proof of concept and quick A/B and usability testing. Based on the insights gathered, I iterated on the design to address pain points, optimize user flows, and improve overall usability, then developed the refined, high-fidelity prototype in Figma ready for stakeholder review and further development.

Other Features

Card Image

Smart Calibration

Analyzes fitting, environment, and user preferences, recommends personalized adjustments tailored to each situation.

Card Image

Personalized Suggestions

Delivers tailored recommendations based on user behavior, extracted data, and lifestyle patterns, ensuring a more personalized experience.

Card Image

Smart Journey

A personalized journey built from the interview results, the user’s lifestyle, shaping an experience reflecting their daily life.

Admin Side Impact

100% complete, structured interviews Higher tuning accuracy Less manual work Consistent data across clinics

User Side Imapcts

Reduced anxiety Faster adaptation Fewer clinic visits Clearer guidance Higher long-term satisfaction

Business Imapcts

Higher conversion Higher retention Lower support cost Long-term personalization capabilities

Tech and Tools
Safety: Human-in-loop approval, conflict detection, completeness scoring

Tech and Code

GitHub

Cursor

n8n

Chatgpt

Supabase

Elevenlabs

Figma Make

MCP Server

8

Lines converging into a central point
Media & Design

Figma

Figjam

Miro

NanoBanana

VEO3

Higgsfield

MidJourney

7

What Is Volta

Project Overview

Volta was an early-stage fitness tech startup aiming to bring social connection back to physical activity. The vision was to let people easily find workout partners, organize matches, book courts, and rent equipment ,all in one seamless mobile experience.

My challenge was to turn a raw business idea into a coherent, user-centered MVP that people would actually use, not another dead fitness app.

I was responsible for the entire UX strategy and product design, from defining the user problems and journeys to building scalable, accessible components and testing usability across flows.

TL;DR

I redesigned Amplifon’s hearing-aid experience by replacing a broken, manual interview workflow and an overwhelming user journey with a two-sided AI-powered platform:
a structured clinical intake system and a voice-first companion app.

This redesign improves accuracy, reduces clinic visits, supports elderly users emotionally, and establishes a long-term personalization backbone across the full hearing lifecycle.

Mobile Image
Mobile Image
Mobile Image

mobin "modaam" piri

mobin "modaam" piri

mobin "modaam" piri

mobin "modaam" piri

©2025 mob! design

©2025 mob! design