FAVFrancisco Álvarez Varas

Projects

Case studies, not just repositories

Each case covers the context, the problem, what I did and the outcome. With links to code and demos where they exist.

AnalyticsJun – Jul 2026

Employee churn prediction on Google Cloud

Anticipate which employees of an engineering consultancy are at risk of leaving, using 24 months of HR records and Slack messages.

Outcome: A reproducible GCP pipeline that processes new data with no manual step, with a retention dashboard in Looker Studio and an interpretable model on Vertex AI.

My role: Ingestion of external labour-market sources, bronze-layer normalisation, retention KPI catalogue and the gold views feeding the dashboard.

BigQueryDataformCloud StorageVertex AILooker Studio+4
Read the case →

AnalyticsJun 2026

Supermarket price monitor

Track Mercadona and Dia prices every day, compare a common basket across stores and see which products go up and which go down.

Outcome: A daily pipeline running on a Raspberry Pi, with price history, a cross-store basket comparison and a trends dashboard.

My role: Full design and build of the pivot, from spec to dashboard: data model, scrapers, product matching, SQL views, CronJob and visualisation.

PythonPostgreSQLSQLKubernetesDocker+3
Read the case →

Data engineeringJan 2026

Air quality monitor for Valencia

Ingest Valencia's air-quality stations in real time, transform the data with dbt and alert when thresholds are exceeded.

Outcome: Recognised as the top team of the cohort. A complete Docker pipeline with dbt every five minutes, Grafana dashboards and Telegram alerts.

My role: Second contributor on the team. Ingestion, dbt transformation models and integration of all services in Docker Compose.

PostgreSQLdbtFastAPIDocker ComposeGrafana+2
Read the case →

AnalyticsApr 2023 – Oct 2024

SEO analytics and experimentation for an e-commerce

Measure which changes on an e-commerce site actually move organic traffic, and automate the reporting management used to receive by hand.

Outcome: 14.3% more organic traffic attributable to the experiments, and KPI reports that generated themselves every week.

My role: SEO and CRO data analyst at GrowBarato, sole owner of analytics for a year and a half.

PythonpandasGoogle Search Console APIGoogle AnalyticsSQL+2
Read the case →

Data engineeringApr 2026

CloudRISK · serverless pipeline for a geolocated game

A Risk over the 87 neighbourhoods of Valencia where troops come from each player's real steps, built as a 100% serverless streaming pipeline on Google Cloud.

Outcome: The full architecture deploys with a single terraform apply. Estimated cost under €3 a month with ten games a day.

My role: FastAPI backend, walker as a Cloud Run Job, Terraform infrastructure, CI/CD and bot simulator. Top contributor to the team repository.

Pub/SubDataflowApache BeamFirestoreBigQuery+6
Read the case →

ProductJul – Sep 2026

Website and booking engine for Hoteles Cumberland

Replace the website of a four-hotel chain with an in-house platform with bookings, an admin panel, a guest CRM and automated emails.

Outcome: In production since August 2026. An in-house booking engine, a hardened admin panel and an automated test suite.

My role: Solo end-to-end project, from proposal to deployment, including an editing manual so the client can maintain the site without coding.

Next.js 16TypeScriptTailwind CSS 4Prisma 7PostgreSQL+5
Read the case →

Data engineeringMay 2026

EDEM Student Hub · virtual campus with an AI agent

A student platform with a FastAPI backend, a conversational agent on Gemini with access to academic data and automated deployment to Cloud Run.

Outcome: Three Cloud Run services deployed with Terraform and GitHub Actions. An agent with 34 tools over 24 Cloud SQL tables and vector search on Vertex AI.

My role: Backend and architecture.

FastAPIJWTGoogle ADKGemini 2.5 FlashCloud SQL+5
Read the case →

Contact

Let's talk

I'm looking for an Analytics Engineer or Data Analyst role in Valencia or remote. Drop me a line and I'll reply the same day.