Fintech Rebuilt Its Data Foundation in 15 Months—AI Never Touched Production
alex_verem · x · 2026-10-08
A PE-backed payments processor hired a team to build a data foundation for AI inside systems handling real transactions under PCI compliance. Before, recurring reports required an experienced data engineer running Lambda functions across 30+ databases and file feeds one at a time; fifteen months later, a business analyst runs them with plain SQL over 1.5B records in one governed warehouse.
Notably, the client wanted no AI near production, and none was deployed there. AI was instead used to map the legacy billing codebase so the access layer could be designed around how applications actually use the data, followed by middleware: 23 data models, table-level permissions, and more. Across dozens of engagements, the author sees teams get this backwards—bolting a model onto scattered data, watching it guess, then declaring AI unfit, when the data work is what makes AI ready.
More from coding & agent
- Dev amazed as AI sets up a fresh computer's config in seconds, not minutes — Dimillian · 2026-10-08
- After a $10,811 Cloudflare bill, dev ditches delayed billing alerts for a Grok-built 3-hour usage monitor — mazzaTalk · 2026-10-08
- Voice agents that call businesses for people: caller ID, disclosure, billing and other pitfalls — Born-Mastodon443 · 2026-10-08
- Harvard study of 718 firms finds AI coding boosts individual output ~30% but firm-level shipping doesn't move — DavidLinthicum · 2026-10-08
- Dimillian shares a 4-step Codex workflow for native-looking iOS apps — Dimillian · 2026-10-08
- WSJ columnist declares 'the year of Linux' is finally here, thanks to AI agents — alexvoica · 2026-10-08