AI/ML Engineer. Scientific foundations.
I develop machine learning and AI systems with a PhD in Physics as my scientific foundation: careful problem framing, reproducible methods, and rigorous evaluation.
My work centers on AI/ML engineering, applied data science, and practical software engineering. I studied physics at the University of The Andes, completed a master's in Complex Systems Modeling, and earned a PhD in Physics Applied to Medicine and Biology at the University of Sao Paulo. My applied industry work overlapped doctoral research, rather than beginning only after it.
At Carta Healthcare, I built ETL and NLP pipelines for clinical data and applied LLM-assisted extraction to registry workflows. The public case study describes responsibilities and workflow boundaries; it does not establish standards conformance or clinical outcomes.
At TripleTen, my work included data-science project review, technical mentorship, live sessions, and debugging support. Clear feedback and communication helped make modeling choices and evaluation assumptions discussable.
I am especially interested in work that sits between research depth and practical execution: machine learning systems, applied AI workflows, and technical communication that makes complex ideas genuinely usable.
I also founded Camargo IT Solutions.
This site presents 5 selected peer-reviewed publications in journals including Chaos and Chaos, Solitons & Fractals, focused on complex systems, nonlinear dynamics, and data-driven modeling. Browse selected publications or view Google Scholar .
Applied work and technical mentorship.
My doctoral research ran from May 2020 to September 2024 and overlapped my applied industry work.
-
Started May 2024
TripleTen
Data Science and AI/ML Instructor
Reviewed data-science projects, delivered technical feedback, and supported learners through live sessions, debugging, and Python/ML instruction.
Read case study -
March 2022 – February 2024
Carta Healthcare
Data Scientist
Built clinical ETL and NLP pipelines and applied LLM-assisted extraction to registry workflows.
Read case study -
February 2019 – December 2021
University of Sao Paulo
Teaching Assistant
Supported university teaching alongside graduate research.
How I work
- Define the problem carefully before reaching for models, prompts, or tooling.
- Keep pipelines reproducible so results can be trusted, debugged, and improved.
- Evaluate claims honestly instead of optimizing for the appearance of capability.
- Make systems understandable enough to trust, especially when the stakes are high.
A quiet place to build.
A small glimpse of the environment behind the writing, systems work, and long technical hours.
