Data Scientist • MLOps & Orchestration • Cloud Data Engineering
Cut Mercedes test-validation time 60% (7.4 h → 2.9 h)
| ML / AI | Data Eng | Cloud | Languages |
|---|---|---|---|
| XGBoost · TensorFlow · PyTorch | Spark · Airflow · dbt · MLflow | AWS · Azure · GCP | Python · SQL · Java |
| Project | Impact | Stack |
|---|---|---|
| Lang2Query | ROUGE-1 ↑ 77% | Mistral-7B · LoRA · RAG · FastAPI |
| CloudBridge | Power BI refresh < 5 min | ADF · Databricks · Synapse |
| SignalSense | Sharpe 1.8 over 5 yrs | XGBoost · Vertex AI · MLflow |
Reduced Mercedes-Benz system test-cycle by 60%, saving ~240 QA hours per release through Random Forest prioritization and PySpark pipeline deployment.
