Fabio Jimenez Olalla

Senior Data Scientist with 5+ years building production fraud detection, anomaly detection, and financial risk models that carry real operational consequence — from AML classifiers and customer behavioral forecasting at Santander SCIB to decisioning pipelines across fintech and enterprise contexts. Comfortable owning model strategy end-to-end in ambiguous environments: defining signals, building from the ground up, and iterating against measurable baselines. Expert in Python, advanced SQL, and translating raw financial data into defensible risk decisions.

Data Scientist | AI Engineer at Inditex Kelea (2026-01 – Present)

• Built production multi-agent data pipeline with role-based access control, automated reporting, and real-time dashboards using FastAPI and Claude SDK; deployed on Azure Containers with OpenTelemetry for full observability.
• Delivered privacy-compliant data access systems and real-time insights at scale with minimal supervision.

Data Scientist at Moeve Bosonit (2024-10 – 2026-01)

• Productionized ML pipelines for workflow automation; owned model evaluation, prompt optimization, and iterative model improvement against measurable baselines.
• Developed end-to-end document classification and lead qualification pipelines; fine-tuned models for specialized scoring tasks and iterated rapidly based on production signal.
• Derived actionable insights from sparse enterprise datasets to drive cross-functional decision-making.

Data Scientist at Permira Bosonit (2023-10 – 2024-10)

• Designed production data pipelines integrating SharePoint (Graph API) with PostgreSQL and FAISS for semantic financial document retrieval and automated analytics.
• Measured business impact of deployed models via experiment design; optimized complex SQL queries across large-scale financial datasets; performed cohort analysis to surface behavioral patterns across customer segments.

Data Scientist AML/Transaction Monitoring at Santander SCIB Bosonit (2021-09 – 2023-09)

• Led ML-based anomaly detection and rule-based classifiers for AML on high-volume SEPA and SWIFT transactional and KYC data under strict privacy and regulatory constraints.
• Applied graph theory to entity deduplication and relationship mapping across transaction networks, enabling detection of anomalous behavioral patterns invisible to record-level models.
• Developed individual and cohort-level behavioral forecasting models to establish expected customer activity baselines, directly feeding downstream anomaly detection and risk scoring.
• Built customer risk profiles using clustering and graph-based modeling; surfaced actionable financial risk signals from sparse, noisy datasets.
• Automated regulatory reporting pipelines and improved risk scoring coverage across diverse customer segments.

M.S. in Business Analytics & Big Data – IMF Business School (2018 – 2019)

B.Eng. in Mechanical Engineering – Zaragoza University (2011 – 2015)