Jasper Venema

• 4 Weeks — Time to Live
• 100+ — Increase in Users per Week for the First 4 Weeks
• 65 — ERE Providers in the Working Prototype

Dutch EV owners and charge point operators can sell emission reduction units (EREs), but the market was opaque: dozens of providers with widely varying commissions, contract terms, and payouts, and no independent comparison. Most consumers picked the first provider they found and left money on the table. The market needed transparency, fast, while the opportunity was unclaimed.

I ran the full product cycle solo. Research: mapped the provider landscape from the official Dutch Emissions Authority (NEa) register, normalizing 65 providers’ commissions, contract terms, and review scores into one comparison model. Build: developed a bilingual (NL/EN) platform with a proprietary earnings engine converting kWh volumes into projected annual ERE earnings with real-time market fees. Delivery: used AI-assisted development (Gemini and agentic coding tools) to ship a production React/TypeScript platform in 4 weeks, with automated register syncing to keep comparisons current.

I achieved live status in 4 weeks with 100+ new users per week within the first month, with zero paid acquisition. Users now compare 90+ providers and get a personalized earnings projection in seconds.

• 20% — Reduction in Data Platform Costs
• 25% — Improvement in Time Efficiency
• 300 — Data Sources Unified in One Data Lake
• 400+ TB — Source Data Processed, Billions of Transactions Daily

Randstad Groep Nederland’s central data lake (300 data sources, 400+ TB, billions of daily transactions) ran on Amazon Redshift, which couldn’t support the mandate to unify all company data: no unstructured data (CVs, video), storage coupled to compute driving unnecessary cost, error-prone manual type mapping, and external cloud sources requiring separate data transport into AWS.

As the product owner of the DataHub team, I led Randstad’s move from Redshift to a decoupled lakehouse: Amazon S3 storage with Apache Iceberg open table formats; Trino/Starburst as the distributed MPP query engine, querying petabyte-scale data directly on S3 and federating lake and operational databases in one SQL query, replacing fragile ETL pipelines; and the open-source DataHub catalog for end-to-end lineage, discovery, and governance. I owned vision, roadmap, and stakeholder alignment, directed specialist partner Riviq on the dual-platform migration, and delivered in 13 months.

I cut platform costs by 20% by decoupling storage from compute and improved time-to-insight by 25% through interactive querying and self-service discovery. I unlocked unstructured data and AI-driven talent matching.

• 3 — New Developers Onboarded/Replaced
• 1 — Head of Product Sourced, Recruited, Hired and Onboarded
• 2 — Quarters of Roadmap and Support Provided

​The client faced significant issues in team structure, product vision, and engineering culture. The fledgling start-up did get into scale-up level agreements and had to reorganize themselves to prepare for future growth, funding, and feature deliverables. The complexity of the domain of systems-engineering and the (then) relatively young field of graph databases ensured a difficult puzzle.

The technical leadership at the scale had been investigated and needed to be replaced. Together with 25Friday we held difficult conversations and provided meaningful research for the suggested organizational changes which where adopted by company leadership. We had to ensure stability and speed of delivery to satisfy investor demands while doing so.

We replaced an underperforming CTO and 2 engineers, hired a head of product, 3 new engineers, and 1 UX designer, and set up the roadmap, feature prioritization, and cost management. We also implemented the first steps in product analytics with Pendo.