OneRagtime

OneRagtime finances and scales early-stage European startups from Seed to Series B. The firm manages €150M in assets, works with 250+ investors, and screens 4,000+ deals per year.

Running a VC firm at that volume needed to be a fully digital system: an investment experience for investors, tooling for every VC operation, and a way to discover and qualify startups and investors without doing it all by hand.

• I designed and built the Platform, the investment experience for investors.
• I built the Core, a digital system that manages all VC operations.
• I built Deepdive, a data pipeline that discovers, qualifies, and ranks startups and investors, from web scraping to AI-driven recommendations.
• I built and managed the technology team (product, development, and data) in Barcelona, and led technical due diligence on startups in the deal flow.

• A Django monolith, PostgreSQL, and scheduled Python scripts for Deepdive instead of an elaborate pipeline

  Why: it worked within weeks, not months

  Cost: it was not elegant.

• Batch processing for Deepdive, matching the initial once-a-day requirement

  Why: simplest thing that met the need

  Cost: when the investment team wanted real-time alerts, retrofitting them onto a batch system was painful.

• Three separate products with strict contracts between them instead of one monolith

  Why: three user groups with different needs could evolve independently

  Cost: more to build up front.

The investment team used Deepdive to make real decisions within weeks of the first version, across 4,000+ deals screened per year. The clean data model made it possible to add new data sources in days instead of weeks. The firm went on to back 50+ companies, including the unicorn Jellysmack.

Docker Swarm · Django · Python · VueJS

Auditing dozens of startups for investors changed how I evaluate technical decisions, including my own. The teams that ship fast aren't the ones with the best stack, they're the ones with the fewest unnecessary abstractions.