top of page

Calibration Intelligence
Workspace (AI-Powered)

Turning complex vehicle calibration data into fast, trustworthy, and collaborative insights.

Automotive | Data Analytics | Embedded Systems

Calibration_Intelligence_Project_Thumbnail.png

🔒  Details in this project are limited due to confidentiality. Let’s connect to explore the story behind the work.

Overview

Calibration Intelligence Workspace was designed as an AI-powered platform to simplify how calibration engineers and data scientists work with complex vehicle testing data. Traditional workflows were highly manual, leading to wasted time, errors in dataset preparation, and siloed collaboration. The vision was to provide a unified tool that validated data, flagged anomalies, and enabled engineers and scientists to collaborate seamlessly, reducing friction while improving decision-making speed.

Challenge

Calibration workflows were fragmented and time-intensive. Engineers struggled with manual data prep, while data scientists needed flexible tools for validation and analysis. Silos slowed collaboration and delayed decisions. The objective was to Design an AI-powered dashboard that helps calibration engineers and data scientists turn complex vehicle testing data into actionable insights.

My Role

UX Lead (Strategy and Interaction Design)

1. Mapped workflows and identified pain points through end-to-end research.

2. Ran interviews and design experiments on novelty detection, smart filtering, and automated reporting.

3. Designed for dual-user needs by balancing simplicity for engineers with flexibility for data scientists.

Impact

1. 30–40% reduction in manual data prep, freeing engineers to focus on analysis.

2. Fewer workflow errors, with metadata validation improving dataset reliability.

3. Unified adoption across engineers and data scientists, breaking silos and speeding collaboration.

4. Insights in hours instead of days, enabling faster, evidence-based decisions.

Reflection

This project highlighted how critical it is to design for two distinct but overlapping user groups. Engineers required speed and simplicity, while data scientists demanded flexibility and control. Creating a platform that satisfied both required careful prioritisation of features and continuous validation through prototyping. A key learning was the value of embedding trust-building mechanisms like metadata validation into the workflow, not only improving data quality but also user confidence. If expanded further, I would explore integrating predictive analytics to surface insights proactively, before issues emerge.

Thanks for reading 🤓

Here are a few more projects that you might find interesting:

Measurement_Data_Management_Project_Thumbnail.png

Calibration Data Management System

Streamlining upload flows and metadata handling across high-volume calibration files.

UX Architecture | Workflow Optimisation | Interaction Design

Field_Service_Ticketing_App_Project_Thumbnail.png

Field Service Technician App

Designed to support field technicians to streamline paperwork and tasks regarding field service documentation.

Interaction Design | User Research

SAPedia_Project_Thumbnail.png

SAPedia: An SAP encyclopedia

Improving enterprise knowledge search for SAP employees.

Interaction Design | User Research | Visual Design

Swarachakra_Project_Thumbnail.png

Swarachakra - A text input application

Usability research for a multi-script text input system

Interaction Design | Gamification | Usability Testing

bottom of page