Adrian Rosari

Pipelines aggregating over 130 sources stabilized. 60 reports migrated without interruption. A 100% manual testing process transformed into an automated tool, reducing PROD incidents.

Freelance Data Analyst Consultant for over 4 years, I work on high-stakes missions where data is critical: banking, insurance, energy.

My profile covers the entire chain: pipeline reliability (SQL, Teradata, PySpark), business process optimization and automation, data governance, predictive and generative AI.

🔹BNP Paribas— Aggregation and stabilization of a critical report on +130 heterogeneous sources (risk, accounting, and local data), documentation entirely reconstructed from scratch without available experts.

🔹AXA France— Migration of 60 multi-region reports (IRIS → Oracle EPM) within deadlines, business continuity maintained without interruption.

🔹Crédit Agricole— Replacement of a 100% manual testing process with an automated tool generating 1,200 test sets: exhaustive coverage of all use cases, elimination of unforeseen production incidents, significant gains in budget and reliability. Regulatory accounting reconciliation established end-to-end.

🔹Enedis— Supervision of 5 flows covering tens of millions of PRMs, detection and resolution of ~15 critical incidents.

I place particular importance on how I work: structured documentation, clean code, clear reporting — because a solution that is not understood or maintained is not a real solution. I also find this requirement as a trainer in Python, Big Data, and AI for engineering and business schools (to name a few: EPITA, EM Normandie, JUNIA-ISEN, EPITECH, NEXA Digital School) — 20 to 40 students per group, ~30 hours per module.

If your data is a construction site, I'm the one for the job.