alfa prep

SZ

Ask Alfa — Driving

AI Agent Adoption

[ Q2 2025 ]

Ask Alfa is an AI-driven insights assistant embedded in Boosted.ai’s investment research platform.

JEFFERSON TOWERS

0%

Product managers
Developers
ML engineers
Sales team

Stakeholders

As a design lead for this project led all design initiatives, worked with a user researcher, designers, product managers, data analysts, developers and UX writers

Company

Role

Ask Alfa is an AI-driven insights assistant embedded in Boosted.ai’s investment research platform

Product

After the initial rollout, engagement metrics were below expectations. Users were hesitant to type prompts, unsure what the AI could do, and rarely returned after the first use.

What was the problem?

_______________

PROBLEM

After the initial rollout, engagement metrics were below expectations. Users were hesitant to type prompts, unsure what the AI could do, and rarely returned after the first use.

Business impact:
Underutilized AI meant reduced ROI on development and missed differentiation in client demos.

Why it was necessary:
Users weren’t confident in how to engage with the AI or what value it could deliver.

Hypothesis

___________________

If we made prompt creation easier and demonstrated the value of AI through relevant examples and suggestions, users would be more confident and engaged — leading to higher adoption and retention.

Ask Alfa is an AI-driven insights assistant embedded in Boosted.ai’s investment research platform

Product

To understand friction points, I conducted user interviews with PMs, analysts, and portfolio managers and reviewed usage data

Findings

Users felt overwhelmed by open-ended text fields
Most were unaware of what data the AI could access
Onboarding lacked connection to real user goals

Methods

Heuristic review of AI onboarding
Journey mapping for first-time users
Analysis of query logs and success rates

Research

To lower the learning curve and build trust, I designed two core UX mechanisms:

2. Prompt Suggestions

Dynamic, context-aware hints appearing directly in chat or data views, teaching users by example without leaving their workflow.

1. Prompt Library

A curated library of ready-to-use, goal-based prompts — e.g.,
“Analyze holdings,” “Compare funds,” “Summarize quarterly performance.”

Each template demonstrated what the AI could do and how to phrase requests effectively.

Design Strategy

To lower the learning curve and build trust, I designed two core UX mechanisms:

2. Prompt Suggestions

Dynamic, context-aware hints appearing directly in chat or data views, teaching users by example without leaving their workflow.

1. Prompt Library

A curated library of ready-to-use, goal-based prompts — e.g.,
“Analyze holdings,” “Compare funds,” “Summarize quarterly performance.”

Each template demonstrated what the AI could do and how to phrase requests effectively.

Design Strategy

[ QUOTE ]

“LIGHT CREATES AMBIENCE AND FEEL OF A PLACE AS WELL AS THE EXPRESSION OF A STRUCTURE”

SWISS-FRENCH ARCHITECT, DESIGNER, PAINTER, URBAN PLANNER, WRITER,
AND PIONEER OF MODERN ARCHITECTURE

1960

LE CORBUSIER

10.11

[ HOUSES ]

AMERICAN ART DECO:
THE SKYSCRAPER STYLE

IN THE LATE 1920S AND EARLY 1930S, AN EXUBERANT AMERICAN VARIANT OF ART DECO APPEARED IN THE CHRYSLER BUILDING, EMPIRE STATE BUILDING AND ROCKEFELLER CENTER IN NEW YORK CITY, AND GUARDIAN BUILDING IN DETROIT

26.12

[EXHIBITION]

POSTWAR MODERNISM IN EUROPE

IN THE LATE 1920S AND EARLY 1930S, AN EXUBERANT AMERICAN VARIANT OF ART DECO APPEARED IN THE CHRYSLER BUILDING, EMPIRE STATE BUILDING AND ROCKEFELLER CENTER IN NEW YORK CITY, AND GUARDIAN BUILDING IN DETROIT

09.01

[EXHIBITION]

LE CORBUSIER AND
THE CITÉ RADIEUSE

SHORTLY AFTER THE WAR, THE FRENCH ARCHITECT LE CORBUSIER WAS COMMISSIONED BY THE FRENCH GOVERNMENT TO CONSTRUCT A NEW APARTMENT BLOCK IN MARSEILLE

MODERN ARCHITECTURE, OR MODERNIST ARCHITECTURE, WAS THE DEFINITIVE ARCHITECTURAL STYLE BASED UPON NEW AND INNOVATIVE TECHNOLOGIES

EVENTS

I owned end-to-end UX delivery, from IA to high-fidelity UI

Deliverables: wireframes, interactive Figma prototypes, and usability test documentation.

Information Architecture

Organized Prompt Library by user roles and workflows

[01]

Interaction Design

Defined logic for contextual prompt surfacing

[03]

Visual Design

Unified with Boosted.ai’s design system

[02]

Collaboration

Worked closely with PMs and ML engineers to align AI behavior and template logic

[04]

ECEXUTION

STAGE 1: BRIEFING

URBAN DESIGN IS THE PROCESS OF DESIGNING AND SHAPING THE PHYSICAL FEATURES OF CITIES, TOWNS, AND VILLAGES. IN CONTRAST TO ARCHITECTURE, WHICH FOCUSES ON THE DESIGN OF INDIVIDUAL BUILDINGS, URBAN DESIGN DEALS WITH THE LARGER SCALE OF GROUPS OF BUILDINGS, STREETS AND PUBLIC SPACES

• COMPUTER ARCHITECTURE 
• BUSINESS ARCHITECTURE
 • COGNITIVE ARCHITECTURE

STAGE 2: PREPARATIONS

INTERIOR ARCHITECTURE IS THE DESIGN OF A SPACE WHICH HAS BEEN CREATED BY STRUCTURAL BOUNDARIES AND THE HUMAN INTERACTION WITHIN THESE BOUNDARIES

10

5

8

[ HOURS ]

[ MEETINGS ]

[ STEPS ]