Mario Ramirez
Data platforms · Machine learning · Systems architecture

Mario Ramirez

Systems Architect & Senior Research Software Engineer

I build data platforms end to end (collection, pipelines, APIs, infrastructure) and turn the data into decisions and peer-reviewed findings. From geophysics to open finance, legal tech and market research; formerly Head of Software & Data Science, now at GESIS and a doctoral researcher at the University of Cologne.

Aachen, Germany GESIS – Leibniz Institute for the Social Sciences

Focus areas

filter projects below
5 / 5

GESIS Surf

Research Software Infrastructure

Open-source research infrastructure for longitudinal web tracking: a browser extension sends windows, tabs, domains, clicks, scrolls and page HTML to a Django REST backend, where Celery processes it, PostgreSQL stores it and Elasticsearch makes it searchable, with ELK monitoring for long-running studies.

Django RESTPostgreSQLCeleryElasticsearchBrowser extension

GESIS Surf Mobile

Research Software

Brings GESIS Surf to smartphones: a cross-platform React Native app that captures how participants browse and use apps, through VPN-based packet inspection on Android and iOS and the Accessibility API, so studies can follow behaviour across desktop and mobile.

React NativeExpoTypeScriptVPN packet inspectionAccessibility API

mcp-server-gesis-surf

Research Software

Connects AI agents to the GESIS Surf infrastructure through the Model Context Protocol: a TypeScript server that lets LLM tools query GESIS Surf's PostgreSQL data and make tightly scoped updates (no inserts or deletes, a WHERE clause required, protected tables, validated inputs). Built with Spec-Driven Development.

TypeScriptMCPPostgreSQLZod

Dynamic Shopper Segments

Modeling & Machine Learning

Doctoral research on behaviour across many retailers: visits are clustered into behavioural states, shoppers are grouped by how they move between those states over time with Markov chains, and Fisher information measures how stable the groups are, separating seasonal swings from real shifts.

PythonClusteringMarkov chainsFisher information

PreisRe

Data Infrastructure

A price-comparison platform for the Mexican grocery retail market — scraping, normalizing, and serving pricing data at scale.

PythonScrapyDjango

Path

Research

Senior Research Software Engineer

GESIS – Leibniz Institute for the Social Sciences · Cologne

I build, design and lead the research software and algorithms that collect and model digital behavioural data from web and mobile, supporting computational social scientists from study design to publication, with open science as the default.

Behavioral analytics

Doctoral Researcher, Behavioral Analytics

University of Cologne · Cologne

Shoppers don't stay in one segment. Using GESIS Surf data, I model how people move between behaviours across retailers, with dynamic clusters and Markov chains, and measure the instability of those moves with Fisher information, to tell seasonal effects from real shifts in behaviour.

Open finance

Senior Full Stack & Data Developer

Finerio Connect · Mexico City (remote)

Reverse-engineered, built and secured data connections to banks' APIs with Python on AWS Lambda, keeping open-finance data accurate, current and protected.

Legal tech

Head of Software Development & Data Science

GeoLegalTech GmbH · Aachen

Joined as a senior full-stack and data science developer, then led the company's software development and data science endeavours in legal tech.

Energy research

Research Associate, Applied Geophysics & Geothermal Energy

E.ON Energy Research Center · Aachen

Monitored geothermal wells with fibre-optic temperature sensing, built the PostgreSQL databases behind it, and turned the data into predictive models and numerical reservoir simulations.

Current research infrastructure

↗
GESIS Surf – longitudinal web tracking
1,070tracked users
>2 TBlongitudinal data
DEGerman extension
Python Django REST PostgreSQL Celery Elasticsearch TypeScript