Back
Bull Technology
FintechB2B0→1

Trading risk tool 0→1 that cut alert response time by 80%

RoleLead Product Designer
ScopeProduct Strategy · UX Research UI Design
TimelineJun 2025 – Present
SHIELD — Dashboard

Impact at a Glance

Faster
Est.80%
Alert response time: 3 min → 12 sec
Faster
Est.78%
New trader onboarding time reduced
Tools
5→1
Unified platform replaced 5 tools

The Problem

What is High-Frequency Trading?

Imagine a trader making 10,000 buy and sell decisions per second — faster than you can blink. That's high-frequency trading (HFT).

The Old Way: Manual Risk Monitoring

Before SHIELD, risk managers monitored trading using a Pre-Trade Risk Management (PTRM) system provided by Borsa Istanbul — Turkey's national stock exchange and the platform that all member trading firms are required to use for pre-trade risk compliance.

The BISTECH PTRM system was designed 15 years ago and never updated for modern trading volumes or team sizes:

BISTECH PTRM
Borsa Istanbul BISTECH PTRM — Figure 12 User Limits / Member Settings Display

The BISTECH PTRM system used by all Borsa Istanbul member firms — dense, fragmented, and built for a different era

Dense spreadsheets with 50+ columns
Alerts buried under noise
No clear visual hierarchy
Response time: 3+ minutes

“In high-frequency trading (HFT), risk can compound in milliseconds. When exposure limits are breached, every second counts. Yet most trading firms manage risk across fragmented tools, with no single view of what's happening across their accounts.”

The Risk Manager's Response Loop

Every second after a limit breach matters. SHIELD was designed around this exact four-step loop — so risk managers can move from alert to action without switching tools or losing context.

A limit is breached — the clock starts
Step 01

Notice

Critical alerts first

Alerts dashboard

severity-sorted

Step 02

Understand

Full context, one click

Detail side panel

who · what · which rule

Step 03

Locate

Straight to the scope

Hierarchy explorer

group · account · user

Step 04

Act

Adjust the limit

Limits editor

usage visible inline

Live order monitoring page

Before & After SHIELD

Before
Trader
Borsa Istanbul Exchange
PTRM
Exchange
5 different monitoring tools
Checking screens every 30 seconds
Alerts buried in tables
3+ minute response time
After
Trader
SHIELD
Borsa Istanbul Exchange
PTRM
Exchange
Nanosecond-speed risk & position checks
1 unified dashboard for the team
Instant critical alerts
12 second response time

Before designing anything, I sat with the people who'd use it.

My biggest advantage was direct access to real users. Our sister company, Liquidity Trading, is an active trading firm working in the exact environment I was designing SHIELD for.

What I did

Click a card to expand

Research Files

Click a file to preview it

SHIELD — research

These insights became the first use cases, which we sketched out and refined with the team.

Figuring out the hierarchy of instruments and accounts was the most critical task — ensuring risk limits were set at the correct scope levels in SHIELD.

Speaking to risk managers helped decide which columns were essential in the main Alerts dashboard, and helped us form early visual concepts quickly.

The alerts table structure we landed on after brainstorming sessions with the team — refined from the early concept into the final column layout.

Collaboration with Engineers

With features being built in parallel, we used a mix of async and sync communication — keeping engineers unblocked between meetings while staying closely aligned when decisions needed a real conversation.

Our Communication Stack

Click any card to expand

Why it worked: Engineers got context without waiting for meetings, and I received technical feedback faster.

SHIELD's Four-Level Hierarchy

Every trading firm has a complex organisational structure. SHIELD mirrors this with a four-level hierarchy so risk managers can set limits at exactly the right scope. Click any level to expand.

Put limits on a specific instrument across all accounts. Useful for high-level risk monitoring.

eg.Set a €5M limit on all EUR/USD trades across the entire firm

Setting limits at an account group level — used when multiple accounts belong to the same trading desk or client.

eg.Limit the London trading desk to $10M in total positions

Account-specific limit enforcement.

eg.Account XYZ cannot exceed $2M in commodity trades

A specific trading user (individual trader login) within an account. Allows user-level limit enforcement.

eg.Trader John can only take positions up to $500K

Solving Complex UX Challenges

Throughout the design process I had to make some key UX decisions. Here are two of the trickiest ones.

Challenge 01

Visualising Map Attributes for an Alert

The Problem

The engineers gave me a SQL sheet which I analysed and found that each alert row can have between 8–10 map attributes. They were too many to add as separate columns, and would be cluttered under one. There was also a possibility of these attributes increasing in future versions.

SQL — Alert Attributes
SQL alert attributes analysis

My Solution — Step 1

I decided not to add them to the alerts table but show all map attributes when the alert row is clicked. This led to a second question...

Accordion or Side Panel?

Should the map attributes be shown in an Accordion or Side Panel when the row is clicked?

Accordion Concept
Early accordion design concept in Figma

Final Decision — Side Panel

Keeping the alert details visible in a side panel while still being able to scroll through other alerts to compare was more important for users than saving horizontal space.

Side Panel
Challenge 02

Too Many Empty Columns on the Positions Page

The Problem

We had one table for markets (top 3 pinned rows) and positions (remaining rows). Some columns like Buy/Sell Quantities weren't relevant for the top 3 rows, and Buy/Sell TL Value weren't relevant for the rest — making horizontal space very inefficient.

Before

Before — cluttered single positions table

After

After — split tables with collapsible markets

My Solution

I split the tables in two — reserving the top table for markets with a collapsible option, making more space for the instruments table below. This made much more efficient use of horizontal space.

Building the Interface

Figma
v0
v0

I made initial designs in Figma, then screenshotted and exported them into v0 for fast prototyping — our features were being built in parallel in the backend, so the team constantly needed to see different prototypes.

Once we agreed on a flow and structure I'd reimport back to Figma to finalise and polish. The frontend developer would use the v0 code and Figma design as a reference to implement.

Building the Interface — SHIELD design file in Figma

Claude Usage for Design System

I connected Claude to Figma MCP to speed up my design process, and created a light mode kit with Claude's help.

SHIELD Design System — Light Mode in Figma

Design Principles

Click any card to see it in action

Why These Colors?

Orange (Primary)

Bull Technology's brand color — carried through into SHIELD as the primary action and highlight color.

Dark Backgrounds

Reduces eye strain during 12-hour trading sessions.

Green / Red

Universal trading signals — profit/loss, safe/danger.

Blue Accents

Trustworthy, calm, professional.

Where is SHIELD Now?

As SHIELD is still not launched, we are currently testing it with our traders at Liquidity Trading for feedback on how we can improve it before the sales team demos it to potential clients.

We're using tools such as HotJar to record traders and use their insights to improve SHIELD's interface before launching this summer 2026.

Launch Target

Summer 2026

Current Testers

Liquidity Trading team

Status

Active user testing & refinement

The Impact

Alert Response Time80% improvement
Before3 min
After12 sec

Measured live with traders — we timed how long it took to spot an alert, open the side panel, identify its source, and begin analysis. Average dropped from 3 minutes on BISTECH PTRM to 12 seconds on SHIELD.

Tools Needed80% improvement
Before5
After1

SHIELD consolidates everything into one platform — real-time alerts, positions tracking, order monitoring, and setting or removing risk limits. Tasks that previously required 3–4 separate tools, each with its own interface to learn.

Time to Train New Traders78% improvement
Before2 weeks
After3 days

After shipping the MVP, we ran a week-long testing session with the trading team at Liquidity Trading, our sister company. When we reconvened, they told us BISTECH PTRM took a couple of weeks to get comfortable with — but with SHIELD's clear navigation and clean UI, they were fully up to speed in 3–4 days.

Reflection

Things I Learnt
How to work better with engineers — basics of CSS, React, Tailwind, and Next.js
How to edit global CSS files from GitHub directly if I want to make changes to implemented pages
UI libraries that developers work with, making it easier to align Dev + Design (Shadcn)
Things I Want to Improve
Creating a design system from the very beginning so there's less back and forth and easier component changes later
Open communication so my team has total transparency into what I'm working on at all times

What's Next for SHIELD

We're constantly improving based on trader feedback. Here's what's coming next.

In Development

AI-Powered Predictions

"Your EUR/USD position might breach in 45 minutes based on current volatility"

Design Phase

Mobile-First Redesign

Manage risk from anywhere, optimised for phone screens

Planned

Custom Alert Rules

"Notify me only if gold positions exceed $1M OR volatility spikes above 15%"

Researching

Multi-Language Support

Supporting global trading teams in their native languages

Check out other projects