Customer Stories

Systems built.
Problems solved.

Three engagements across financial services, government, and SaaS. Purpose-built systems that replaced manual processes with intelligent, reliable infrastructure.

Financial ServicesAI / Document Intelligence

Balaton Financial — AI-Assisted Mortgage Operations

Mortgage operations running entirely on spreadsheets and email — 4 hours per application, no audit trail, brokers spending half their day chasing documents.

Live System Demo Balaton Financial — Mortgage Operations Platform
Active mortgage pipeline — 21 files
18 On track3 Attention
New Enquiry 4
Alex M.
$720k · PAYG · First home
AI Score 91 — Strong
Jordan P.
$1.2m · Self-Employed
AI Score 62 — Review
Docs Collecting 8
Sam K.
$450k · PAYG
4/5 docs
Auto-chase sent 09:40
Maya R.
$890k · Self-Employed
2/7 docs
3rd auto-chase sent
AI Assessment 5
Chris L.
$580k · PAYG
AI: APPROVE — 8 min total
was 4hr manual review
Priya S.
$620k · PAYG
Compliance: PASS
Submitted 4
Taylor B.
$650k · PAYG
Awaiting lender
CRM auto-updated
Document AI — Sam K. file — all docs classified automatically in under 2 seconds Model accuracy: 94.1%
PDF
payslip_march_2026.pdf
Auto-classified 0.8s · Income $142,000 p.a. extracted
PAYSLIP0.97
Verified
PDF
bank_statement_cba_q1.pdf
3 months · Avg monthly expenses $3,640 extracted
BANK STMT0.94
Verified
JPG
drivers_licence_front.jpg
Identity confirmed · Expiry valid · OCR complete
PHOTO ID0.99
Verified
PDF
tax_return_2025.pdf
Requested auto · 2nd chase sent 45 min later · no broker action needed
AWAITING—
Pending
Verified income
$142k p.a.
Expense ratio
0.31 — LOW
AI recommendation
APPROVE
Total processing time
8 minutes
Before vs. After — measurable impact for the brokerage 31 hours saved this week
Before
Time per application4+ hours
Document chasingManual emails
Compliance checksManual review
CRM updatesManual entry
Audit trailNone
Files per broker/week~6
After
Time per application8 minutes
Document chasingFully automated
Compliance checksAI — instant
CRM updatesReal-time sync
Audit trail100% coverage
Files per broker/week~30 (5x)
System Architecture
Broker Portal ─▶ Azure Doc Intelligence ─▶ ML Classifier
│
Pipedrive CRM ◄─ FastAPI Backend ◄─ Compliance Engine
│
PostgreSQL + Audit Log
8 min
Average processing time (was 4+ hours)
94%
ML classification accuracy on income documents
100%
Audit trail coverage — zero compliance gaps
PythonTensorFlowscikit-learn FastAPIPostgreSQLAzure Doc Intelligence Pipedrive APIReactDocker
Government TechnologyNLP / Data Engineering

Trovio — Government Tender Intelligence

No centralised view of government project records — 8 portals checked manually every morning, match scoring done by eye, bid responses drafted from scratch each time.

Live System Demo Trovio — Government Tender Intelligence Platform
8 portals monitored — 14 new opportunities this week
4 High match6 Review
TenderAgencyValueAI MatchDeadlineStatus
TND-1831Dept. of Infrastructure$240k943 daysDraft in progress
TND-1856ATO$85k8810 daysReady
TND-1902DFAT$1.2m6117 daysReview
TND-1915NDIS$320k3224 daysLow match — skip
Portals monitored
8 continuous
Time to identify
Automated
Was: daily manual check
2–3 hours/day
Win rate improvement
+40%
AI Analysis — TND-1831 — Dept. of Infrastructure GPT-4 summary ready
Extracted requirements
Data Engineering Python Azure PostgreSQL GovERP integration Privacy Act compliance 12-month contract Security clearance req.
Capability match score
Data Engineering
95%
Azure Cloud
90%
Govt compliance
85%
Overall match
94%
GPT-4 generated summary

The Department of Infrastructure requires a contractor to design and implement a data integration pipeline connecting legacy procurement databases to a new Azure analytics platform. Python ETL, PostgreSQL schema design, and Privacy Act 1988 compliance are mandatory.

High match — recommend pursuing. Deadline 3 days.

Draft bid sections auto-generated from capability library. Estimated 2h to complete response vs. 12h from scratch.

Deadline tracker — zero missed submissions since go-live 2 deadlines this week
3 days
TND-1831 · Dept. of Infrastructure · $240k
Draft in progress · 2 sections remaining · auto-reminder sent to team
10 days
TND-1856 · ATO · $85k
Ready to submit · internal review complete · one click to lodge
17 days
TND-1902 · DFAT · $1.2m
Under review · large contract · assign team member prompted
Skipped
TND-1915 · NDIS · $320k
AI match score 32 — auto-flagged as low priority, no action needed
System Architecture
Govt Portals (x8) ─▶ Web Scraper ─▶ NLP Pipeline
│
React Dashboard ◄─ FastAPI + PostgreSQL ◄─ OpenAI Summariser
8
Portals monitored continuously
GPT-4
Automated summarisation of tender documents
60%
Reduction in opportunity identification time
PythonspaCyNLTK OpenAI APIFastAPIPostgreSQL AzureReact
SaaSData EngineeringDigital Marketing Analytics

Unleash Live — AI Events, CRM & Marketing Intelligence

AI-generated crowd intelligence from live events had no structured pipeline into CRM or analytics. Marketing campaigns ran on stale data. SEM/SEO workflows were manual. Insights that should have driven decisions were lost the moment they were generated.

Real-time data pipeline CRM automation Marketing analytics SEM/SEO automation Executive dashboards
Live System Demo Unleash Live — Data, CRM & Marketing Intelligence
Live ETL pipeline — events flowing in real-time from 4 venues All systems nominal · 28s lag
Before the pipeline
AI event dataLost immediately
CRM accuracyStale / manual
Executive reportingWeekly, manual
Insights acted onDays later
After the pipeline
AI event dataCaptured & stored
CRM accuracyReal-time sync
Executive reportingLive dashboards
Insights acted on<30 seconds
Pipeline stages — all running
Unleash Live ingestwebhook · real-timeRunning
ETL normaliserdedup + enrich + validateRunning
dbt transforms15 models · 45 tests passingRunning
Salesforce syncREST API · 30s lagRunning
Tableau refreshextract · every 5 minRunning
Salesforce CRM — AI event summaries synced automatically in real-time 38 records updated last batch
Account / VenueAI SignalCrowd DensityAction TriggeredSynced
ANZ StadiumHigh density — NW quadrant87%Alert sent to ops team10:15:09
ICC SydneyNormal dispersal42%No action — nominal10:14:38
Allianz ArenaGate B bottleneck forming64%Task created in Salesforce10:13:52
Qudos Bank ArenaPost-event dispersal18%Event closed — log archived10:12:11
CRM sync lag
<30 seconds
Records today
1,847 updated
Sync errors
0
Was: manual updates
Days delayed
Executive dashboard — live data, updated every 5 minutes automatically Last refresh: 10:15:12
Events today
28,442
+12% vs avg
CRM records updated
1,847
Automated
Pipeline lag
28s
SLA <60s
dbt test pass rate
100%
45/45 tests
Tableau refreshes
96
Today
Manual work today
0 min
Fully automated
ANZ Stadium
8,421 events
ICC Sydney
6,218 events
Allianz Arena
4,941 events
Qudos Bank
3,674 events
SEM & SEO automation — campaigns updated from live event data All automations running
Campaign performance — this week
Google Search — Venue AI
Auto-updated from event data · 3x/day
+34% CTR -18% CPC
LinkedIn — Enterprise
Audience synced from Salesforce pipeline
+21% conv. Stable CPL
SEO content pipeline
Event insights auto-published to blog API
+12 articles Auto
Retargeting — past attendees
Audience from event attendance data
+48% ROAS Live sync
Before vs. after automation
Before
Campaign updatesManual, weekly
Audience segmentsStatic CSV uploads
Reporting4hrs/week manual
Event data utilised0%
After
Campaign updatesAuto, 3x/day
Audience segmentsLive CRM sync
ReportingReal-time dashboard
Event data utilised100%
System Architecture
Unleash Live API ─▶ ETL Pipeline ─▶ Salesforce CRM
│
Tableau Dashboards ◄─ dbt transforms
<30s
CRM sync lag from AI event stream
3
Platforms unified — Salesforce, Tableau, SEM tools
100%
Marketing pipeline automated — zero manual exports
PythonSalesforce APITableau dbtPostgreSQLREST APIsFastAPI

Additional customer stories available under NDA. All client data and architecture details remain strictly confidential.

Ready to build something that works?

Start with a description of your problem. I'll respond within one business day with an initial assessment.

Get a Quote