KavuthamrajSenior UI/UX & Product Designer

Steadfast · AI service-delivery platform

AI does the work. People stay in charge.

Software companies juggle five tools to win a client and deliver an app. Steadfast puts all five steps in one place: AI does the typing and the maths, and a person checks every step.

  • Only designer
  • Founder talks to handoff
  • Shipped
Steadfast Vortex: the lead list with stat cards and AI winning scoresSteadfast Turing Sync: the AI asks a person which database to use
5
AI apps on one shell
1
designer, founder talks to handoff
Shipped
built and used, from my designs
Role
Lead product designer, the only designer
Company
Liffto Technologies
Industry
IT services, AI SaaS
Platform
Web app, desktop
Services
IA, UX, UI, design system

The story in 30 seconds

Think of a company that builds apps for other businesses.

To win one client and deliver one app, its team goes through five steps:

  1. Step 1Win the clientVortex
  2. Step 2Write down what to buildAxiom
  3. Step 3Put a price on itPrism
  4. Step 4Plan the workCadence
  5. Step 5Build the appTuring

Before

  • Five separate tools
  • The same details typed again at every step
  • Nobody noticed when the AI got stuck

With Steadfast

  • One place for all five steps
  • AI does the typing and the maths
  • A person checks and approves every AI step

Four decisions, in 90 seconds

Short on time? These four choices shaped the whole product. Each one links to where you can see it.

What was going wrong

Every client went through five tools, and the same details were typed again at each one.

  • Typing everything by handClient details copied from emails, PDFs and calls.
  • Plans full of holesMissing and repeated requirements.
  • Prices hidden in spreadsheetsRebuilt for every client. Only one person understood the maths.
  • AI stuck, nobody knewThe AI waited for answers no one could see.

The idea

Let AI do the boring work, and let a person see, fix and approve every step.

The five steps, screen by screen

One app per step, all in one place. Each shows what changed for the person using it.

Module 01 · Vortex

Win the client: a new lead in seconds

Sales pastes an email or a PDF. The AI fills in the form and says how likely the deal is to win.

BeforeTyped by hand from emails and callsNowAI fills it from what sales already has

Try it: add a lead with AI

Prompt

Brief from Northwind Pay, a fintech company: a payments app for small shops, about 1.5M. Contact Priya Raman, Head of Product. Proposal due 30 May.

northwind-rfp.pdf

Waiting for the AI0%

Lead details

Rebuilt in code with made-up example data. Fields the AI fills are marked; edit one and it becomes yours. Nothing counts until a person submits.
Steadfast: The AI score explains itself: strengths, weaknesses, next best actions and the factors behind the number (opens in new tab)
The AI score explains itself: strengths, weaknesses, next best actions and the factors behind the number

Module 02 · Axiom

Write down what to build

The AI turns rough notes into a proper plan. The analyst approves or rejects every change.

BeforeDrafts with gaps and duplicatesNowAI tidies them and points out what's missing

Steadfast: Pick a template, and the AI sorts the notes into it (opens in new tab)
Pick a template, and the AI sorts the notes into it
Steadfast: Edit the plan with an AI assistant beside it (opens in new tab)
Edit the plan with an AI assistant beside it

Module 03 · Prism

Put a price on it

No more hidden spreadsheet maths. Every hour and every cost is a line anyone can read.

BeforeRebuilt in a spreadsheet for every clientNowA clear price list anyone can read

Try the estimation model

Grand total 1,370 h

Engineering effort 1,000 h
Buffers 100 h
Project overheads 240 h
Agile ceremonies 30 h
Example numbers, not a real deal. Buffers and overheads are a share of engineering effort, ceremonies scale with sprints, and any line can be switched off.
Steadfast: Studio sizing: design effort by screen type and complexity (opens in new tab)
Studio sizing: design effort by screen type and complexity
Steadfast: Proposal studio: AI drafts, people edit (opens in new tab)
Proposal studio: AI drafts, people edit

Module 04 · Cadence

Plan the work with AI teammates

AI helpers and people share one task board. Every task shows who owns it right now.

BeforeWork planned by handNowAI drafts the plan, the manager adjusts it

Click through a task's life

PLTX-12Unassigned

As a user I need to share dashboards with clients
  • UI Screens
  • API
OwnerNobody yet: AI assigns it by default

New from the backlog. Unless a person steps in, the AI picks the owner, human or agent.

The task flow from the requirements, rebuilt in code. Each state names who owns the task.
Steadfast: AI agents and people on one team (opens in new tab)
AI agents and people on one team
Steadfast: One sprint board for both (opens in new tab)
One sprint board for both

Module 05 · Turing

Build it, and ask when stuck

When the AI writing the code can't decide, it asks a person, and the answer box is right there.

BeforeNobody sees when the AI is stuckNowThe AI asks, and a person answers in one place

Steadfast: Turing Sync: messages grouped by stage with health dots, and the AI's question answered in place (opens in new tab)
Turing Sync: messages grouped by stage with health dots, and the AI's question answered in place

From a spreadsheet row to a real screen

Every screen started as a row in the requirements spreadsheet. Follow one feature through.

ModuleSub-moduleFeaturePriority
PrismOverviewEstimation, P&L, planning and proposal toolP1
PrismLandingSparks in a Kanban or list view; a click opens the sparkThis row became the screenP1
PrismSummaryEvery detail of the selected proposalP1
Wireframe of the Prism Sparks list in greyscale
Final Steadfast design of the Prism Sparks list

It starts as one row in the requirements spreadsheet.

ModuleSub-moduleFeaturePriority
CadenceKanbanBacklog, Sprint 1, Sprint 2… as columnsThis row became the screenP1
CadencePlan with AIAI spreads tasks by priority and effortP1
CadenceKanbanEach sprint column shows its ticket countP1
Wireframe of the Cadence sprint board in greyscale
Final Steadfast design of the Cadence sprint board

It starts as three rows in the requirements spreadsheet.

What if the AI gets it wrong?

It will, sometimes. So every mistake is easy to spot, cheap to fix and never final without a person.

  1. 1It fills a field wrongFields stay editable until submit.
  2. 2It suggests a bad changeAdd or Cancel. No silent edits.
  3. 3Its score looks arbitraryReasons and factors are shown.
  4. 4It hits a call it shouldn't makeIt asks, then waits for review.
  5. 5The work is too riskyEscalated to a person.
  6. 6Who changed what?Every action is logged.
One screen, rebuilt in code with made-up example data. Hover or tap a number.
See the full processResearch, personas, structure, flows and the design system.Open itClose

How I worked

  1. 01Understand

    Founders, requirements, competitors

  2. 02Define

    Journey, goals, IA

  3. 03Design

    Wireframes to final UI

  4. 04Systemise

    One shell, one system

  • Sales leadWorks in Vortex

    Goal
    Log leads fast
    Pain
    Retyping RFPs and notes
    AI
    Autofills and scores leads
  • Business analystWorks in Axiom

    Goal
    A complete SRS, first time
    Pain
    Gappy, duplicate drafts
    AI
    Maps drafts, flags gaps
  • Delivery managerWorks in Cadence

    Goal
    Plan and track delivery
    Pain
    Blocked AI work is invisible
    AI
    Drafts plans, flags human tasks
Proto-personas from founder sessions and the requirements.

What it had to do

  • No retypingEach step reuses the last.
  • AI you can checkVisible, editable, signed off.
  • No hidden waitingEvery wait is seen and counted.

The shell

Five apps, plus three shared tools. One place to switch.

Each app gets a letter tile and a colour, and the colour follows it through the product. Pick a tile.

Vortex

AI-assisted sales: capture leads, assign them, track stages and score the chance of winning.

Designed in this case study
Steadfast shellGoogle sign-in · app switcher · notifications · profile
  • Vortex
    • Spark list and board
    • Add a lead with AI
    • Spark detail and AI score
    • Firms, peers, insights
  • Axiom
    • Sparks and SRS status
    • Template library
    • Map a draft
    • Studio and AI review
  • Prism
    • Summary
    • Studio and dev sizing
    • Estimation and pricing
    • Proposal studio
  • Cadence
    • Projects with RAG
    • Team: AI and people
    • Sprints and tasks
    • Gantt, P&L, risk
  • Turing
    • Setup and boilerplate
    • UI listing
    • Turing Sync
    • Docs and vault
Information architecture. One shell, then a short menu per module; each opens on a list of deals or projects.

Add a lead with AI

  1. StartOpens Add Lead
  2. PersonAdds a prompt, voice note or RFP
  3. AIFills the lead details
  4. DecisionAll fields right?NoEdit the wrong one
  5. PersonSubmits the lead
  6. AIScores it and explains why
  7. EndLead joins the list
The AI does the typing; a person checks and submits.

The AI needs a person

  1. StartAI hits a hard call
  2. AIFlags it: needs a human
  3. AIAsks in Turing Sync
  4. PersonAnswers in place
  5. DecisionAnswer enough?NoAI follows up
  6. EndAI resumes, logged
Waiting on a person is a visible, counted state.

Plan a sprint with AI

  1. StartOpens Sprints
  2. PersonSets length, team and dates
  3. AIPlans by priority and effort
  4. DecisionPlan looks right?NoAdjust, AI re-plans
  5. EndSprint starts
The AI drafts the plan in the open; the manager can step in.

Design system

  • Primary#605BFF
  • Pressed#3D38E3
  • Ink#151338
  • Primary tint#ECEBFF
  • App background#F5F7FA
  • Axiom#B16FF3
  • Prism#26C0E2
  • Cadence#FB8050
  • Turing#86C270
  • Error, high priority#F15E5E

AaPoppinsRegular 400 · Medium 500 · SemiBold 600

  • 32 px · 600AI winning score 90%Page titles
  • 20 px · 400AI winning score 90%Section titles
  • 16 px · 500AI winning score 90%Body and table text
  • 14 px · 500AI winning score 90%Labels and buttons
  • 12 px · 600AI winning score 90%Tags and captions

Primary button

Add LeadDefault
Add LeadHover
Add LeadPressed
Add LeadDisabled

Secondary button

Add LeadDefault
Add LeadHover
Add LeadPressed
Add LeadDisabled

Text field

Client nameEnter nameDefault
Client nameNorthwind PayHover and focus
Client nameNorthwind PayFilled
Client nameClient name is requiredError

Status and score

In ProgressStatus
CompletedStatus
Ready To SubmitOutline status
90%AI winning score

Tabs

Lead DetailsDocumentsAI ScoringActive tab: 2 px primary underline on a 1 px baseline

Search

SearchSearch
NorthwindSearch, active
Rebuilt in code from Steadfast's tokens: 10 px corners and 48 px filled fields, #605BFF with #7470FF hover, #3D38E3 pressed and #ACABBF disabled.

What I delivered and learned

What I handed over, and what I'd do differently next time.

What I delivered

  • The structure and key flows for five apps
  • Wireframes and final UI with every state
  • One design system for every app
  • A shipped product, built from my designs

What I learned

Design the waiting
Progress matters as much as the answer.
Give the AI a face
Named agents read as teammates.
Test with users early
Next time, sales and delivery staff before handoff.
Check colour as text
Cyan and green need darker text tones.
What I'd build next
A confidence level per field, linked to the sentence it came from.

Client names and figures in the screens are placeholder data, not real clients or results.

→ Hiring a product designer?

Let's talk about your product.

Open to Senior Product Designer, Product Designer and UI/UX roles. Immediate joiner, open to relocation or remote work.

Next case studyERP Solution for Sports Industry