We build, ship and manage custom data solutions.
Pipelines, lakehouses, and the apps and AI agents that run on them. We prove the numbers are right, then keep it all running in production.
Open any project and use the app.
Each case study starts with the running product, with its numbers beside it.
Offline-first inventory for TT Global
Store staff scan QR codes and record stock on a phone, even when the network drops.
Pharmacy data feed migration
A 165-column patient status and dispense feed moved from SQL Server stored procedures to Databricks, and proved row for row against the legacy output before cut-over.
All work
- 01Meta4 PaymentsAutomated quotes and payment links: from a WhatsApp chat to a paid invoice, with agentic AI drafting the quote and code deciding the money.Full-stack products2026→
- 02Merciven: revenue-leakage detectionA live SaaS that reads an SME's own spreadsheets and finds where money is leaking, with an AI layer that can't make up numbers.Applied AI2024→
- 03LegacyOS: oral knowledge, searchableElders' recordings transcribed with Whisper and made answerable, with every answer citing who said it and when.Applied AI2025→
- 04Offline-first inventory for TT GlobalStore staff scan QR codes and record stock on a phone, even when the network drops.Full-stack productsDelivered→
- 05Pharmacy data feed migrationA 165-column patient status and dispense feed moved from SQL Server stored procedures to Databricks, and proved row for row against the legacy output before cut-over.Data engineering2026→
- 06Hospital data platformIngestion, masking and dbt transformation for six US hospitals on Databricks and Snowflake.Data engineering2026→
- 07Sociolytiks backendThe backend of an AI social media management SaaS: scheduling, AI content, approvals and paid tiers.Full-stack products2025→
- 08Truth estimation and research searchAn API that scores how likely a statement is to be true, using a research corpus, and shows its evidence.Applied AI2023→
- 09Testing money-moving featuresRelease validation for a fintech app, and a sample suite that checks the rules a bank cannot break.Automation, QA & immersive2026→
- 10ÌMỌ̀: LLM agents that know LagosTwo LLM agents that share one persona engine: one writes the review a Lagos diner would leave, the other ranks places for them and explains why.Applied AI2026→
- 11Visual defect detectionAn autoencoder that learns what a good product looks like, rejects anything it can't rebuild, and watches for drift in production.Applied AI2025→
- 12Network traffic analyzerReads packet captures, shows who is using the network and how, flags port scans and spikes, and suggests what to fix.Data engineering2024→
- 13Blush BudgetA personal budget app that plans the month, logs spending and works out the flags and savings tips from your own numbers.Full-stack products2026→
- 14XR showroom with gaze analyticsA 3D gallery you can walk through in the browser or a VR headset, measuring live which exhibits hold your attention.Automation, QA & immersive2024→
Four practices, one team accountable for each delivery.
Full-stack products
Web applications with payments, authentication and real users, from database to interface.
02Applied AI
LLM features that are grounded and tested: the model drafts, typed rules and computed results decide.
03Data engineering
Pipelines, lakehouses and dbt models on Databricks, Snowflake and Azure, including healthcare data.
04Automation, QA & immersive
Test suites for money-moving systems, operations automation, and data from XR experiences.
The model drafts. Typed rules and computed results decide.
AI that can't invent numbers
Language models propose; code checks. Figures come from calculations, and any AI sentence that cites something that doesn't exist is rejected.
Proven against what you have
New pipelines are compared with the old system column by column before anyone switches over.
Money handled like money
Verified webhooks, idempotent ingestion, row locks and integer kobo. Retries and concurrent payments are tested, not hoped about.
You talk to the people who build it
No hand-offs. Based in Lagos (UTC+1): a full day shared with the UK and Europe, and your morning in the US.
Have something to build or fix?
Tell us about it on a free 30-minute call. You'll leave with a clear next step, whether or not you work with us.