Data Scientist · Buenos Aires
Hi, I’m Ignacio Morales.
I turn messy data into reliable models
and decision-ready data products.
End-to-end builds. Open one to read the writeup and try the live demo where there is one.
Loan Default Fairness
An end-to-end credit model whose decision rule maximizes profit, not accuracy — and a disparate-impact audit that prices what fairness would cost.
Buenos Aires Real Estate
59,069 for-sale listings from two sources, scraped and cleaned into one dataset — then mapped to median price per m² across all 48 barrios of the city.
Checkout Conversion A/B Test
A pre-registered A/B test on 286,690 users — power analysis, an SRM check, and a confidence interval that makes “do not ship” as rigorous a call as any win.
I’m a Data Scientist based in Buenos Aires, Argentina.
I work with data end to end: collecting it, structuring it, and turning it into something people can actually use. That means building and calibrating machine learning models, getting usable datasets out of messy real-world sources, and running experiments that help separate real effects from noise.
Right now, I build and operate a production system that ingests external data every day and flags entities that behave differently from their own historical baseline. A big part of my work is the less glamorous side of data: reconciling records across sources, checking what doesn’t add up, and finding inconsistencies before they make it into a report.
I care about the parts that are easy to overlook. Making sure a model’s probabilities actually mean something. Designing an experiment with enough power to trust the result. Checking whether a finding matters in practice, rather than stopping at statistical significance. The underlying principle is the same: let the data lead instead of forcing a story onto it.
I see accuracy, a clean chart, or a positive result as a starting point, not the finish line. I’m drawn to problems where the answer isn’t obvious yet, and I work in both Spanish and English.
Toolchain
Python