Über Maximilià
I'm a physicist who ended up in data. Same instinct, different tools: figuring out how complex systems actually work.
Over 8 years I've helped companies turn data into systems that drive real decisions, working across the full data chain: modelling, pipelines, dashboards, experimentation, machine learning and applied AI. I'm particularly drawn to regression modelling and interpretability, breaking KPIs down to identify which variables genuinely move them and by how much, so teams know which levers to pull.
Recent work:
- Adevinta (Kleinanzeigen): rebuilt CRM notifications data infrastructure across one of Europe's largest classifieds.
- United Nations (WFP / ShareTheMeal): built whole data landscape, A/B testing framework with 50+ experiments, propensity models in production at 85–96% accuracy, and ML for hunger forecasting.
- Zalando: 2 years across demand, pricing, warehouse and merchandising.
What I typically deliver:
- Data foundations: modelling, governance, pipelines (AWS, Redshift, BigQuery, Databricks) and source-of-truth datasets.
- BI & reporting: Tableau dashboards designed around decisions, not vanity metrics.
- Experimentation: A/B testing frameworks from design to infrastructure.
- Machine learning: propensity, forecasting, segmentation, clustering. Deployed in production, not stuck in notebooks.
- Research & deep-dives: KPI driver analysis, customer behaviour, attribution, funnels.
- AI & automation: LLM-powered text structuring, RAG, internal assistants.
I work with teams at any level of data maturity — from companies still building foundations to those with mature stacks looking for the next layer of impact. The first step is always the same: assess the current state, identify what would create most impact from there, and execute.
Based in Ibiza, fully remote across Europe, available to travel on-site when it adds value. Fluent in English, Spanish, Catalan and Italian.
Englisch
Muttersprachlich oder zweisprachig
Spanisch
Muttersprachlich oder zweisprachig
Italienisch
Muttersprachlich oder zweisprachig
Katalanisch
Muttersprachlich oder zweisprachig
Französisch
Konversationssicher
Portugiesisch
Konversationssicher
Deutsch
Konversationssicher
Nur remote
Führt Projekte hauptsächlich remote aus
Projekt- und Berufserfahrung
- Adevinta - KleinanzeigenFreelance Data ConsultantE-COMMERCEJuli 2025 - Dezember 2025 (5 Monate)Barcelona, SpanienLed the end-to-end analytical definition of the CRM notifications domain (push, email, in-app) at Kleinanzeigen, one of Europe's largest classifieds platforms, taking the team from zero visibility to a complete understanding of the messaging funnel and user engagement.
- Designed the full data model: raw events, fact and model layers (sent → delivered → opened → session) and a behaviour-enriched mart layer; and defined the KPI framework, event taxonomy and naming conventions powering all downstream reporting.
- Performed extensive data discovery and exploration in Databricks: profiled event quality, audited inconsistencies across CRM systems and surfaced tracking gaps in the underlying database.
- Partnered closely with a data engineer on the production data model and with CRM engineering, Product and Design to co-define new events aligned with the analytical schema.
- Built Tableau dashboards covering the full notification funnel, KPIs and user engagement, iterating with business stakeholders to ensure clarity, usability and consistent visual language.
- Conducted deep-dive analyses on messaging logic, send timing and frequency, notification types, opt-in/opt-out behaviour and Notification Center usage, surfacing actionable insights for CRM optimisation.
- Translated complex analyses into clear narratives and stakeholder-facing decks, enabling data-driven decisions on CRM strategy and laying the foundation for future tracking and optimisation work.
- World Food Programme - UNData AnalystHIGHTECHNovember 2021 - Juli 2025 (3 Jahre und 8 Monate)Berlin, GermanyLed the analytical development of ShareTheMeal and wfp.org for the United Nations World Food Programme (WFP), the world's largest humanitarian organisation fighting hunger and a 2020 Nobel Peace Prize laureate. Designed pipelines and analytical layers in AWS / Redshift, productionised ML models and LLM applications in Python, and built Tableau dashboards to enhance decision-making across Marketing, Product, CRM, Finance, Content and Management.
- Built the end-to-end reporting layer for ShareTheMeal across all business verticals (Marketing, Product, CRM, Finance, Content, Management), establishing a single analytical foundation for the organisation.
- Designed and ran a complete A/B testing framework enabling 50+ experiments, driving measurable improvements in conversion and retention.
- Developed and deployed propensity and revenue projection models (85–96% accuracy) integrated into CRM campaigns and business dashboards (+12% upsell increase), plus market-clustering models segmenting countries to tailor in-app prompts and messaging (avg. +6% conversion).
- Ran regression models to forecast critical hunger metrics across three countries, providing leadership with data-driven foresight for resource allocation.
- Consolidated multi-platform attribution data into a single source of truth, integrating feeds from multiple ad platforms and Adjust, and engineered an alternative iOS attribution pipeline working around SKAdNetwork limitations.
- Structured raw BigQuery event data into a unified sessions table powering full-funnel visibility of onsite user behaviour across web and app.
- Implemented automated text-structuring flows using LLMs, turning customer support chats, reviews and survey responses into actionable insights.
- ZalandoData AnalystE-COMMERCEOktober 2019 - November 2021 (2 Jahre und 1 Monat)Berlin, GermanyJoined Zalando, Europe's leading online fashion and lifestyle retailer, through a one-year rotational programme across Demand, Product, Warehouse and Pricing, before moving permanently into the Sales & Markets team at Zalando Lounge, the group's flash-sales platform for off-season fashion.
- Built the stock source-of-truth for Zalando Lounge: designed AWS/Redshift pipelines consolidating fragmented stock feeds into a single reliable view used across merchandising and operations.
- Designed an offer vs. demand ML score combining product-type and user conversion rates against available stock, helping merchandisers decide weekly assortment and balance returns.
- Developed campaign performance dashboards for merchandisers (brands, themes, weekly cadence), partnering with stakeholders to align on KPIs and reporting needs.
- Conducted deep-dive analyses on country and customer-cluster behaviour, online user journeys (funnels, session times) in BigQuery, and quantified category contribution and the impact of delivery delays on cancellations through regression modelling.
- Performed warehouse operational analysis profiling processing times and volumes by station, market and brand.
- Prototyped a dynamic discount-rate model in Pricing aimed at optimising weekly revenue margin.
Empfehlungen
Sei die erste Person, die Maximilià empfiehlt
Teile Deine Erfahrung aus der Zusammenarbeit mit diesem Freelancer.
Diese Freelancer passen auch zu Ihren Kriterien
Agatha Frydrych
Backend Java Software Engineer
4.7
(3)
2
Baptiste Duhen
Fullstack developer
4.6
(4)
5
Amed Hamou
Senior Lead Developer
4
(2)
7
Audrey Champion
Web developer
4.3
(3)
4
Ausbildung und Abschlüsse
- Data Science DiplomaUbiqum Code Academy2019Data Science Diploma
- MSc in Renewable EnergiesUniversidad Autónoma de Madrid2013MSc in Renewable Energies