I design and build end-to-end data solutions with deep Data Engineering roots: SQL pipelines, data warehouses, and automated reporting that turn raw operational data into decisions organizations can rely on. Work amplified with Agentic AI (Claude, Copilot, Cursor) to ship faster without giving up rigor, including this site itself.
Senior BI Engineer with deep Data Engineering roots, work amplified with Agentic AI. I design and build end-to-end data solutions: SQL pipelines, data warehousing, and automated reporting that guarantee information integrity and accessibility.
Most recently at CTT — Correios de Portugal, where I authored the #1 and #2 most-viewed Power BI reports across the entire company (and most of the top 10 overall), supporting decision-making across 20+ departments and 500+ stakeholders, including executive leadership. Day-to-day toolkit: Power BI, SQL Server, Azure Databricks (including Databricks SQL) and Power Automate; earlier in the same role, PySpark on the medallion pipelines. Years of analysing event-level operational data across a national distribution network: sequence, stage-by-stage drop-off, SLA per hop.
When I onboard a new domain, I tend to map it onto the previous one at the data-model level. At CTT, postal logistics became the electrical distribution network I knew from EDP: sorting centers as substations, mailmen as meters. That isn't a metaphor. Both industries produce event and state data on a redundant distribution network, and both ask the same questions of it: uptime, incidents, loss per node.
My work is about helping people do more fulfilling tasks by automating information delivery: taking repetitive reporting off their plates so attention can go to decisions and judgment.
I do my best work in async environments that value autonomy: outcomes over activity, writing over meetings, building over reacting.
And I care about the next person who'll read the work: consistent nomenclature, readable schemas, decisions documented inline.
Same logic, applied to people: direct conversations over polite ambiguity, clearer for everyone, faster to converge.
CTT — Correios de Portugal
National postal operator
key results
USERPRINCIPALNAME(), coordinated with site authentication. 10,000+ users at time of build.Audiogest / PassMúsica
National music rights society
key results
Multivision (project at Nokia)
key results
EDP Distribuição
National electricity distribution grid
Postgraduate in Information Management
Specialization in Business Intelligence & Knowledge Management
NOVA IMS · Lisbon · 2019 — 2020
Master's Degree in Marketing
Universidade do Algarve · Faro · 2010 — 2015
Bachelor's Degree in Hospitality Management
Universidade do Algarve · Faro · 2007 — 2010
# Used across professional work, personal projects, and continuous exploration
A growing collection of case studies, dashboards, and data tools.
Architected embedded Power BI Row-Level Security for a multi-tenant delivery-quality data product. One dataset, scoped per authenticated client via USERPRINCIPALNAME(), serving ~10,000+ B2B clients through their customer site. 5+ years in production, no known leaks.
A daily-fresh operations-exception dashboard for a national postal and logistics operator. Built when the platform had no orchestrator, no writable database, and no native partial-refresh in Power BI. Four years in production, consistently #1 or #2 most-viewed report across the organisation.
The site you're reading. A meta-project on shipping a portfolio in a single afternoon by pairing with an AI agent: what I scoped, what I delegated, and the honest tradeoffs.
Single-page tool for the Switch 2 game Pokémon Pokopia. Ranks habitat items by how many of your roster's Pokémon would care, surfacing the actionable answer at a glance. 307 Pokémon, 607 items, 43 categories, with a vanilla JS frontend fed by a self-healing Python scraper.
Open to conversations about data, BI strategy, or interesting problems. Reach out through any of the channels below.