Gabriela Pozo
What the data says, and what it doesn't yet.
Open-data analysis: a question, a finding with its source and what could not be measured. I worked on indicator tracking and foresight studies in the public sector, and I'm motivated to use evidence to decide better and create public value.
- Education
- Wildfires
- AI and official data
- Specialty coffee
- Tree cover
Read the projects GitHub Email Download CV (PDF, in Spanish)
Projects
Open-data cases. Each one poses a question, presents the finding and puts the source one click away.

School cohorts in Peru 2025: which districts to serve first?
151 of 1,798 districts are high priority
Question
Which districts will face the most pressure or gaps in the school system over the next 6 years, and which should be served first?
Finding
151 of 1,798 districts are high priority (151,020 children aged 3 to 5, 11.0% of the country); 44% are in Loreto, Huánuco and San Martín, and 36 districts have a triple gap. This is not an official projection.

Wildfires in Peru: what the satellite could measure
1,690 ha burned in an area with no hotspots recorded between 2012 and 2025
Question
How much burned in Peru's largest fires in October 2026?
Finding
In Chichas (Arequipa), 1,690 ha burned, 4.3% of the district, in an area with no heat hotspot recorded between 2012 and 2025, which rules out recurring managed burning. Only 4 of the 20 largest fires could be measured; clouds covered the rest.

Does Claude make up figures? Memory, web and the BCRP's MCP
10/10 correct answers with the BCRP MCP server (web: 9, memory: 0)
Question
Does Claude answer official BCRP data correctly depending on where it gets the information?
Finding
With 10 questions and Claude Haiku, the MCP server got 10 right, the web 9 and memory 0 (it said “I don’t know”). When memory was denied the “I don’t know” option, 7 of 10 answers came out wrong: 3 were real estimates and 4 a zero used as a placeholder.
More cases
FICAFE archive: ten editions of the specialty coffee fair
Ten editions of Peru's International Specialty Coffee Fair (2016-2026) with linked sources, and what remains unmeasured.
Exploratory analysis with open data
Analysis notebooks with open data on the fiscal crisis, global energy consumption and tree cover loss in Peru.
Contact
Do you have a question that data could answer? Write to me and let's talk.