AI/ML Engineer · Data Scientist · Mathematical Optimization
AI Ops Specialist @ IBM Deutschland R&D · Ehningen, Germany
I build production AI systems at IBM — RAG-based assistants, agentic AI workflows, and mathematical optimization models for enterprise SAP and cloud deployments. My background in Mathematical Engineering bridges deep statistical foundations with real infrastructure engineering: I work where the math meets the system.
Currently shipping:
- 🤖 RAG pipelines & agentic AI on IBM Cloud for enterprise SAP workloads
- 📐 Optimization models for cloud virtualization pricing and capacity planning
- 🔭 AI observability environments for on-prem and cloud infrastructure
Languages
ML / AI
Data & Stats
Cloud & Infrastructure
My current production work at IBM is internal. The repositories below are from my Mathematical Engineering degree at EAFIT University and my collaboration with Hospital Universitario de la Sabana — they cover the research foundations I apply daily in production.
Patient Assignment from Game Theory and ABMS
Agent-based simulation of nurse–patient workload dynamics in a hospital inpatient setting. Implements game theory assignment mechanisms within a multi-agent model to analyze workload distribution and emergent staffing patterns. Built in collaboration with Hospital Universitario de la Sabana.
PythonMesaAgent-based modelingGame theoryHealthcare systems
Implementations of core deep learning architectures from scratch and with TensorFlow/Scikit-learn: MLPs, CNNs (LeNet-5), GANs, autoencoders, and unsupervised clustering. Covers the full supervised → generative → unsupervised spectrum.
PythonTensorFlowScikit-learnCNNsGANsDeep learning
SIR epidemiological compartmental models applied to Colombia's early pandemic data. Uses curve fitting and least-squares optimization to calibrate model parameters against observed case trajectories.
PythonSciPyEpidemiological modelingOptimizationCurve fitting
Mathematical Engineer · EAFIT University (2019–2023) · Spanish (native) · English (C2) · German (B1)