Writing / Series
Case Studies
Real deployments, real numbers. What we actually built, who it was for, and what changed once it shipped.
Everything else on this site is argument and blueprint. These are receipts: systems Cone Red built and put into production, with the before-and-after numbers attached. A bank compliance stack that went from rule-based to 99.6% recall. A recommendation engine running in 100+ pharmacy locations off a 2M-relation medical graph. A one-button voice app that ended a four-person paper relay in a Spanish clay quarry.
No case study here ships without the failure modes and the constraints that made it hard: language barriers, offline environments, legacy systems nobody wanted to touch. If you build with AI for a living, this is the part where the model meets a business that has to keep running while you change it.
- How We Built an AML System for 21 Banks: From Rules to AI Anti-money-laundering compliance used to mean rule-based systems that caught 80% of threats and flooded analysts with false positives. Here is how we replaced that stack with AI across 21 banks processing $250M+, and what we learned shipping it.
- Pharmacy AI Helper: How We Built a Recommendation Engine for 100+ Pharmacy Locations Pharmacists cannot memorize thousands of products. We built an AI system that scans any product and instantly returns alternatives, complementary items, and treatment protocols, powered by 2M+ medical relations. Here is how it works and what it changed.
- From Paper Chaos to Voice AI: How We Automated a €2.3M Clay Mining Operation Every truck arrival at a Spanish clay quarry triggered a 4-person paper relay across language barriers. We replaced it with a one-button voice app. Here is how we built it and what it changed.
