Reconciliation and settlement data: ledgers, payouts and card networks
I am Syed Ahmed Basharat Ali, a data scientist and analytics engineer in New York City. I have worked on settlement and reconciliation data at Geidea, Saudi Arabia's largest payments processor, and built reconciliation from zero for a multi-tenant warehouse as a consultant.
I have also published public projects on why common reconciliation checks miss the errors that cost the most.
Settlement and reconciliation at Geidea (Aug 2023 to Jan 2025)
- Built settlement and onboarding data models covering 400K+ POS terminals.
- Built an automated Tableau platform for FinOps reconciliation, SLA tracking and interchange optimization across 5+ teams: 40% less manual reporting and 60% faster Oracle dashboards.
- Built real-time ML anomaly detection in Talend ETL on 1M+ daily payment transactions to catch fraud and settlement errors.
- Segmented 40K+ merchants by MCC, interchange and chargeback behavior.
Warehouse reconciliation at Rime (consulting client)
Through my own consulting practice I was the CEO's data person at Rime (rime.sa), a Saudi edge-AI retail camera startup. I shipped an 11-model dbt migration across a multi-tenant Snowflake warehouse (50+ client schemas) with reconciliation and observability built from zero in Python and GitHub Actions. It surfaced two real sync issues on its first run.
Public work on ledgers, payouts and card data
- Ledger integrity under agent-speed traffic: Netting hid $32,732 (46%) of $71,570 gross error; a retry with a regenerated idempotency key was caught 0 of 53 times and is 93% of undetected dollar-hours.
- Marketplace payout reconciliation: 122,810 orders; daily checks fire 31 of 31 days on a clean book; a fee-on-tax bug passes all reconciliation until fees are recomputed from policy.
- AI spend intelligence from a raw card feed: 231 card descriptors resolved to 15 vendors; metered spend is not callable before about day 12 of the month, and is plus or minus 100% at day 5.
What these show
Daily-total reconciliation sees the net error, not the gross: in ledger-integrity, netting hid 46% of the gross error. Reconciliation proves that systems agree with each other, not that they are right: in marketplace-payout-reconciliation, a fee computed on tax passed every check until fees were recomputed from policy.
Tools
SQL, Python, Oracle, Talend, Tableau, dbt, Snowflake, PostgreSQL, GitHub Actions.
Contact
Email sabasharat.ali@gmail.com, or find me on LinkedIn and GitHub. Full history on the experience page.
Open to roles in New York or remote US.