About TendiosTendios is the tender intelligence platform for the Spanish public procurement market. Companies use Bid to find, qualify and win public tenders. Public institutions use Create to draft and manage their own. Behind both products is one of the most complete datasets on Spanish public procurement: tenders, lots, CPVs, contracting bodies, resolutions and awards. It’s collected continuously from hundreds of public sources and enriched with AI.
That data is our product. We’re looking for the person who will own it end to end.
The roleThis is not a classic data architect role. You won’t just design schemas and hand them off. You’ll own Tendios data from the moment it’s scraped to the moment a customer sees it in Bid, asks Vera about it, or queries it through our MCP server. You’ll write the code, propose the features, and make the calls on how data is modelled, stored and exposed, including how it feeds our AI features.
You’ll report directly to the Head of Technology and work closely with product, the AI & Data squad and the backend squads.
Propose and build data-driven features. Examples: market and competitor intelligence from award data, pricing benchmarks, contracting-body profiles, better tender matching and alerting, and data-quality signals shown to customers.
Own the data side of our AI features. Design and improve the retrieval layer behind Vera and our Dynamic RAG service: document ingestion and conversion (Docvert), chunking, embeddings, Qdrant indexing, hybrid search with Elasticsearch, and reranking.
Use LLMs where they add real value in the pipeline: structured extraction from tender documents, classification (e.g. CPVs), entity resolution and summarisation. Build evaluation sets to measure quality instead of guessing.
Work with the AI team so Vera and the Tendios MCP server get clean, well‑structured,



well‑documented data and tools.
Define and measure data and retrieval quality: coverage per source, freshness, parsing accuracy, deduplication and RAG answer quality. Treat it as a product metric, not an afterthought.
Senior Data EngineerRemote (Spain) or hybrid from BarcelonaFull‑timeAbout TendiosTendios is the tender intelligence platform for the Spanish public procurement market. Companies use Bid to find, qualify and win public tenders. Public institutions use Create to draft and manage their own. Behind both products is one of the most complete datasets on Spanish public procurement: tenders, lots, CPVs, contracting bodies, resolutions and awards. It’s collected continuously from hundreds of public sources and enriched with AI.
That data is our product. We’re looking for the person who will own it end to end.
The roleThis is not a classic data architect role. You won’t just design schemas and hand them off. You’ll own Tendios data from the moment it’s scraped to the moment a customer sees it in Bid, asks Vera about it, or queries it through our MCP server. You’ll write the code, propose the features, and make the calls on how data is modelled, stored and exposed, including how it feeds our AI features.
You’ll report directly to the Head of Technology and work closely with product, the AI & Data squad and the backend squads.
What you’ll doPropose and build data-driven features. Examples:



market and competitor intelligence from award data, pricing benchmarks, contracting-body profiles, better tender matching and alerting, and data-quality signals shown to customers.
Own the data side of our AI features. Design and improve the retrieval layer behind Vera and our Dynamic RAG service: document ingestion and conversion (Docvert), chunking, embeddings, Qdrant indexing, hybrid search with Elasticsearch, and reranking.
Use LLMs where they add real value in the pipeline: structured extraction from tender documents, classification (e.g. CPVs), entity resolution and summarisation. Build evaluation sets to measure quality instead of guessing.
Work with the AI team so Vera and the Tendios MCP server get clean, well‑structured, well‑documented data and tools.
Define and measure data and retrieval quality: coverage per source, freshness, parsing accuracy, deduplication and RAG answer quality. Treat it as a product metric, not an afterthought.
Own the data platformTake ownership of the collection pipeline (Crawl Manager, Tenders Discovery, Collector, Tenders and Resolution Parsers, RabbitMQ consumers) and make it more reliable, observable and cheaper to run..
Keep our read models consistent with the system of record: Elasticsearch for search and filtering, and Qdrant for semantic search and RAG.
Build out the analytics layer on ClickHouse, orchestrated with Prefect and exposed through Metabase. This includes replacing our manual, Excel‑based SaaS metrics reporting (MRR, churn, NRR) with a proper warehouse.
Write and ship codeWrite production code in Python for the pipeline and data services and in TypeScript/Node.js in our Turborepo backend (NestJS) where data features touch the API.
Review code, set standards for data m

📌 Senior Data Engineer (Barcelona)
🏢 Tendios
📍 Barcelona