Services

I help teams design and ship applied AI and data systems — from RAG assistants and research datasets to real-time environmental monitoring. Below are the main areas I work on, grounded in production and research experience.

What I offer

Conversational AI & RAG systems

Domain-specific assistants grounded in your documents, with local or private LLM options and production-ready APIs.

  • RAG over technical or operational documentation
  • User-trained / on-premise LLM setups
  • FastAPI backends and workflow automation (e.g. n8n)

Dataset research, evaluation & creation

Research-oriented work on building, curating, and evaluating datasets for NLP and applied AI — not only training models.

  • Corpus design and dataset creation for research or shared tasks
  • Data quality evaluation and suitability analysis
  • Evaluation pipelines (e.g. hallucination / labeling consistency) for research teams

AI architecture for monitoring platforms

Architecture for multi-site sensor and alert systems — ingestion, analytics hooks, and early-warning flows.

  • End-to-end AI/data architecture for environmental or hydraulic monitoring
  • Real-time ingestion and alerting design
  • Requirements and system design for multi-city deployments

IoT & meteorological data platforms

Platforms that collect, store, and visualize high-frequency meteorological and IoT sensor data.

  • TimescaleDB / time-series storage and ETL
  • REST APIs and operational dashboards
  • Station networks and board-style operator views

Featured project

SGM — Meteorological dashboards (San Andrés Island)

San Andrés Island, Colombia · Corporación Del Laboratorio al Campo · EEDAS

Software Engineer

Meteorological monitoring system to track weather variables across stations on San Andrés Island. Operators select station dashboards (e.g. Brooks Hill, RSU, Interaseo) for recent observations and board-style views used in day-to-day monitoring.

TimescaleDB · REST APIs · Streamlit / web dashboards · IoT ETL

Open live dashboards

Interested in working together?

Tell me about your use case — NLP, monitoring systems, or data platforms.

Contact