Designing an AI companion that guides people from raw data to a confident decision.

Overview

Product

Fosfor AI, an AI companion built into the Fosfor Decision Cloud.

Problem

Users across very different roles struggled to navigate complex data workflows and pull out insight they could act on.

Details

UI and UX design across research, architecture, and prototyping over twelve weeks.

Solution

An AI guided companion that supports three personas while keeping data secure and trusted.

Result

A single intelligent layer across the whole data to decision journey.

Impact

3

Personas served, Data Designer, Insight Designer, and Decision Designer

12 weeks

From strategy to a tested prototype

4

Layers in the companion architecture, context, interaction, response, and memory

Context

Fosfor is LTIMindtree's data and decision platform. Fosfor AI is the intelligent companion built into it, meant to help people interact with data and reach decisions without fighting the tools.

The hard part was that one companion had to serve three very different people, the one preparing data, the one finding insight in it, and the one making the call, each with their own needs and their own idea of what help looks like.

Problem

Across all three roles, people hit the same wall, complex data workflows and no easy way to extract insight they could act on. Each persona struggled at a different point, and none of them had intelligent assistance that fit how they actually worked.

Goals

One companion that makes every step from data to decision feel guided rather than fought.

  1. 01

    Boost real productivity across the data to decision journey

  2. 02

    Fit the experience to three distinct personas

  3. 03

    Build trust and security into every interaction

  4. 04

    Keep the assistance contextual, not generic

Process

Research and insights

Mapped the needs of all three personas through interviews and analysis.

Ideation and wireframing

Explored how an AI companion could sit inside existing workflows without disrupting them.

Prototyping and testing

Built interactive prototypes and refined them, including split testing of key interactions.

Key insights

  • 01

    Data Designers needed help managing and transforming large datasets before anything else.

  • 02

    Insight Designers needed sharper tools to analyse data and derive insight from it.

  • 03

    Decision Designers needed an intuitive way to navigate recommendations and act with confidence.

Solution

Fosfor AI, a single companion that adapts its help to whoever is in the data to decision seat.

Design

Contextual intelligence

Suggestions and prompts grounded in the user's current task, not generic AI noise.

Design

A unified experience

One consistent companion across the three personas, so help feels the same wherever you are in the journey.

Design

Security and trust

Clear controls and transparent operations, so people trust the companion with their data.

Design

The companion architecture

A four layer model, context, interaction, response, and memory, that lets the assistant hold a thread across a session.

A selection of companion and workflow screens.

Before and after

Before

After

The workflow before and after the companion, the same task made guided.

Good AI assistance does not show off, it quietly removes the next obstacle in front of the person.

My design principle for Fosfor AI

Reflection

Designing for three personas at once taught me to find the shared spine of an experience before tailoring the edges.

If I were taking it further, I would measure whether the companion actually shortened the path from data to decision in real use, which is the result the project still needs.

From raw data to a confident decision, with a companion the whole way.