Data Science · AI Engineering · Data Engineering
I build applied machine learning, data engineering, and AI agent systems with an emphasis on reproducibility, evaluation, automation, and evidence-based engineering.
| Project | Focus |
|---|---|
| DevFlow Engineering Analytics | Data engineering pipeline for GitHub analytics with typed contracts, 16 deterministic data-quality rules, reproducible CI, and 155 automated tests. |
| FieldOps AI Agent | Applied AI field-operations PoC integrating MCP tools, Microsoft Copilot Studio, and Telegram with authenticated and idempotent workflows. |
| Telecom Churn Prediction | Reproducible churn modeling pipeline using training-only cross-validation, an untouched final holdout, and risk segmentation. |
| Asuna ML Agent | Public ML systems case study with leakage-aware screening, training-only model selection, and reproducible scoring workflows. |
| Brújula Vocacional | Governed Colombian vocational knowledge base and evaluation specification for bounded RAG and conversational agents. |
| Interactive Portfolio | Next.js portfolio with a context-bounded AI assistant powered through Cloudflare Workers AI. |
Data & ML
Python · SQL · pandas · scikit-learn · Machine Learning · Data Quality · ETL/ELT
AI Systems
AI Agents · RAG · Model Context Protocol (MCP) · Microsoft Copilot Studio · Microsoft Foundry
Engineering
Git · GitHub Actions · TypeScript · Node.js · CI/CD
Cloud & Analytics
Microsoft Azure · AWS · Power BI · Microsoft Power Platform
Building stronger applied data, ML, and AI systems with reproducible experimentation, reliable evaluation, and clear engineering boundaries.
