Mariana Fonseca
Data Scientist, Marketing and Product Analytics
Porto, Portugal · UTC+0 · 2 years of experience
Mariana focuses on turning messy data into tested models and useful decisions.
Request a shortlistWhy this profile fits
Decision-first problem framing
Starts with the decision, target metric, and data limits before choosing a model or analytical method.
Validated modeling
Uses holdouts, experiment design, and error analysis to separate a useful signal from a convincing notebook.
Clear operational handoff
Documents assumptions, monitoring needs, and the path from analysis to a decision or production workflow.
Relevant toolkit
PythonSQL
scikit-learn
PyTorch
dbtDelivery evidence
Selected work
Churn Early-Warning System
Subscription D2C commerceBuilt a gradient-boosted churn model on 180k subscribers that scored every account weekly and fed a targeted retention experiment. The at-risk cohort it surfaced drove a win-back campaign that reduced voluntary churn by 14% over eight weeks, with the pipeline productionized in dbt and MLflow.
Experimentation Framework and A/B Test Library
Consumer mobile app product teamCreated a reusable Python library and Streamlit review app for designing and reading A/B and holdout tests, with built-in power analysis and pre-registered hypotheses. Standardizing significance testing cut hasty false-positive launch calls and gave product managers a consistent way to judge results.
Relevant experience
Career timeline
Data Scientist, Marketing and Product Analytics
2024–PresentDevlyn (client engagements) · Remote, Porto, Portugal
- Built and shipped a gradient-boosted churn model on 180k plus subscribers for a subscription commerce client, feeding a weekly retention experiment that cut voluntary churn by 14%.
Junior Data Analyst
2023–2024Lumo Subscriptions · Porto, Portugal
- Owned weekly growth reporting across acquisition, activation, and retention for a D2C brand with 300k plus monthly active customers, in SQL and Looker Studio.
Data Science Intern
2023–2023Meridian Growth Studio · Remote, Lisbon, Portugal
- Supported analytics for six client accounts, building recurring dashboards and ad-hoc analyses on campaign performance across Google and Meta.
Hiring this role?
See how Mariana fits the Data Scientist role.
Skill depth
Show experience in context.
Relevant experience by skill
See the complete skill index
Languages
Libraries/APIs
Frameworks
Tools
Platforms
Storage
Paradigms
Other
Education
Formal background
MSc, Data Science and Advanced Analytics
NOVA IMS, Universidade NOVA de Lisboa · 2022–2024 · Concluded with distinction, 16 of 20
BSc, Economics
Faculdade de Economia, Universidade do Porto · 2019–2022
Credentials
Certifications
dbt Analytics Engineering Certification
dbt Labs · 2024 · Certified
Communication
Languages
Portuguese
Native
English
Professional
Spanish
Conversational
More examples
Similar Data Scientist profiles.
Rohan Iyer
Senior Data Scientist
Senior data scientist, 13 years turning demand and sales signals into forecasts operations teams actually act on.
PythonSQLTime-Series ForecastingDemand Forecasting+8
Marta Wójcik
Senior Data Scientist
Senior data scientist, 11 years, focused on churn and retention modeling.
Camila Andrade Rocha
Senior Data Scientist
Senior data scientist with 9 years turning behavioral data into models and decisions, specializing in experimentation and causal inference.
PythonSQL+8