I build production-oriented software that connects backend engineering, data pipelines, machine learning, and AI automation. I have 7+ years of professional experience modernizing business systems, developing ERP and CRM platforms, automating data-intensive workflows, and applying machine learning to classification, segmentation, forecasting, and decision support.

Backend: C# · .NET · ASP.NET Core · REST APIs · WPF
Data & ML: Python · SQL · pandas · NumPy · scikit-learn · LightGBM · XGBoost · CatBoost · TensorFlow · Keras · Optuna
Databases: PostgreSQL · pgvector · MySQL · SQL Server · SQLite
MLOps & Infrastructure: FastAPI · Apache Airflow · MLflow · Docker · Redis · Terraform · GitHub Actions
AI: RAG · LLM workflows · model evaluation · grounded retrieval · AI-assisted automation

Status: Currently under private development. The repository will be linked after it is prepared for public release.

Problem: SaaS sales workflows require reliable lead prioritization, explainable predictions, secure knowledge retrieval, and reproducible model delivery.

Action: Designing and implementing a Docker-first platform that integrates FastAPI, PostgreSQL/pgvector, MLflow, Airflow, Redis, Terraform, gradient-boosting models, Optuna tuning, SHAP explanations, DBSCAN segmentation, monitoring, release controls, and grounded RAG.

Current progress: Established the platform architecture and implemented core foundations for data processing, model development, retrieval, orchestration, monitoring, security, and controlled releases. Model-performance metrics are intentionally omitted until the project is ready for public review.

Python FastAPI PostgreSQL pgvector Airflow MLflow Redis Docker Terraform LightGBM XGBoost CatBoost Optuna SHAP

Problem: ASP.NET Core applications need authentication and authorization infrastructure that separates security behavior from database-specific persistence.

Action: Developed a provider-neutral .NET 10 authentication package architecture with password authentication, registration, refresh tokens, OIDC support, diagnostics, host integration, and dedicated relational providers.

Result: Established supported SQLite and PostgreSQL integration paths, exact-revision verification gates, reproducible PowerShell-based validation, and a controlled release process for public package delivery.

C# .NET 10 ASP.NET Core PostgreSQL SQLite OIDC Authentication Authorization PowerShell

Problem: Local AI assistants often require hardware resources that make private, CPU-first deployment impractical.

Action: Defined a phased architecture for a low-resource BitNet assistant, including monorepo boundaries, PostgreSQL/pgvector services, hardware discovery, model-manifest validation, local tooling, documentation, and CI scaffolding.

Current result: Completed the governed Phase 0 foundation while keeping chat, RAG, model serving, authentication, orchestration, and MLflow experiments behind explicit implementation gates.

Python BitNet PostgreSQL pgvector RAG uv pnpm CI/CD

• Modernized a legacy DOS/BASIC and dBASE application by designing and developing a C# and MySQL ERP that integrated onboarding, point of sale, billing, CRM, and domain-specific operational modules.
• Developed Python machine-learning solutions for forecasting, classification, clustering, segmentation, anomaly detection, and operational pattern analysis.
• Automated email classification, response generation, product-restocking workflows, data extraction, and reporting through Python, n8n, SQL, Excel, and VBA.
• Converted unstructured and operational data into validated datasets, model-ready features, KPI reports, and ranking logic for business decision support.

I prioritize:

• Clear architecture and explicit interfaces
• Reproducible experiments and leakage-safe validation
• Automated testing, static analysis, and measurable quality gates
• Security-aware API and authentication design
• Honest documentation of implemented capabilities, limitations, and pending verification
• Maintainable solutions with controlled algorithmic and operational complexity

• Computer Software Engineering coursework — National University of the South
• Generative AI Leader - Google Cloud Certificate through National University of Cordoba
• Google Data Analytics Professional Certificate — Google
• Programming with Python Professional Certificate — OpenEDG Python Institute
• Career Essentials in Generative AI — Microsoft and LinkedIn
• English: C1 Advanced — EF SET

I am interested in opportunities involving Machine Learning Engineering, Backend Software Engineering, AI Engineering, and Data Engineering.

• LinkedIn: linkedin.com/in/jonatancordoba
• GitHub: github.com/JonatanCordoba
• Email: jonatan.cordoba.1991@gmail.com