Case studies
Platforms delivered.
Lessons from operation.
Read the constraints, engineering decisions and outcomes from our platform, data and operations work. Client identities are omitted where the engagement is confidential.
Alongside these delivery accounts, our tenant-scoping audit method explains how we review isolation across a shared platform.
- Anonymised delivery Sports & Gaming
Consolidating many wagering tenants onto one platform
Tier-one wagering operator
Every tenant ran on its own infrastructure, with its own administration and its own reporting. Cost scaled linearly with tenant count and nobody could see consolidated risk.
Platform to Platform · 24×7 Operations Desk NSW · Victoria · Northern Territory - Anonymised delivery Sports & Gaming
A multi-tenant tote and fixed-odds platform, built and operated
Multi-tenant tote and fixed-odds platform we built and operate
An operator wanting to launch a wagering brand needs racing content from several providers, a wallet, a tote and fixed-odds bet path, an agent hierarchy and a settlement model — and does not want to run its own stack to get them.
- Build period
- 2020–2025
- New tenant deployment
- Minutes
Platform to Platform · Data to Data · 24×7 Operations Desk NSW · Victoria · Northern Territory - Anonymised delivery Sports & Gaming
Real-time probability modelling for win and exotic pools
Racing analytics platform we built and operate
Racing data arrives from inconsistent sources at different times in different shapes, and every new predictive factor a data scientist wants to test has to be built by an engineer first — so the modelling cycle runs at engineering speed, not research speed.
- Pools modelled
- Win and exotic
- Factor authoring
- Custom DSL
- Production history
- Several years
Data to Data · Platform to Platform NSW · Victoria · Northern Territory - Anonymised delivery Sports & Gaming
A tipping app that sized stakes and placed bets on the exchange
Mobile tipping product we built and later sold
Tipping apps hand the user a selection and stop there. The user still has to decide how much to stake, open another app, find the market, and get on before the price moves — which is where most of the value of a good tip is lost.
- Mobile platforms
- iOS + Android
- Risk profiles
- 3
- Execution
- Automated
Data to Data · Platform to Platform Other - Anonymised delivery Retail & FMCG
Pricing 100+ dairy SKUs against volatile milk input costs
Large dairy manufacturer
Product costs were re-estimated manually against a raw material whose price moves with fat content, quality, season and supply, so pricing decisions across a 100+ SKU catalogue were made on stale cost assumptions.
- SKUs modelled
- 100+
- Margin improvement
- 5-12%
- Forecast accuracy
- >90%
Data to Data · Platform to Platform Other - Anonymised delivery Government & Civic
Reading public sentiment by ward from four civic data channels
Government agency, reached through an enterprise partner
A government agency had four separate streams of citizen opinion — news coverage, complaints, feedback forms and phone surveys — and no way to turn any of them into a comparable read on sentiment for a given ward.
- Classification accuracy
- 92%
- Reporting granularity
- Ward level
Data to Data · Platform to Platform India - Anonymised delivery Government & Civic
Identifying individual tigers from live video by stripe pattern
State government wildlife authority in India
Forest staff identified individual tigers by eye from camera footage. The work was continuous, the volume of footage exceeded the number of people available to watch it, and sightings were logged after the fact rather than tracked as they happened.
- Identification accuracy
- 96%
- Surveillance workload
- ~70% lower
- Labelled training images
- 20,000+
Data to Data · Platform to Platform India - Anonymised delivery Sports & Gaming
Timer-driven top-ups were quietly manufacturing promo liability
A live-games module on the multi-tenant wagering platform we build and operate
A live-games module topped up its shelf of live bets on a timer, and every quiet period converted that fresh stock into promo liability the moment each race jumped.
Platform to Platform Other - Anonymised delivery Sports & Gaming
A stuck verification is a stuck customer: operating a KYC pipeline
The identity-verification pipeline of a multi-tenant platform we build and operate
Verification state lived in three places — our pipeline, the third-party provider, and the truth — and whenever they disagreed, a customer sat stuck at a deposit or withdrawal gate with no automated way out.
24×7 Operations Desk · Platform to Platform Other - Anonymised delivery Sports & Gaming
Schema-migration crash loops at deploy, and the tooling that ends them
The multi-tenant tote and fixed-odds wagering platform we build and operate
Rolling deploys raced additive ScyllaDB schema migrations, so new pods expecting columns the live schema did not yet have crash-looped on startup and held the rollout hostage.
24×7 Operations Desk · Platform to Platform Other - Anonymised delivery Financial & Insurance
Two kinds of out-of-memory, and why telling them apart is the fix
A multi-currency ledger service on the multi-tenant platform we build and operate
A production JVM ledger service experienced both Java heap exhaustion and container memory-limit failures. Each required different evidence and remediation.
24×7 Operations Desk Other
Why most of these are anonymous
An NDA is a normal condition of this work, not an evasion.
Wagering platform contracts routinely forbid naming the operator or the brand running on the platform. We honour that, and we would honour it for you.
-
What we will always say
What was built, what it integrates with, what the sequence was, and what we would do differently. The engineering is the part you need.
-
What we will not say
The operator, the brand built on the platform, or any descriptor narrow enough to identify one company. A thinly-veiled hint is the same disclosure with extra steps.
-
What that means for you
Your engagement gets the same treatment by default. If you later want to be named, that is a decision you make in writing, not one we assume.
Want a reference conversation?
There is a limit to what a public page can say about NDA work. What we can do is talk you through the architecture and the failure modes in detail.