Wizenoze

The ed-tech scale-up whose technology I built from scratch as CTO: a search and content platform that crawled the open web, classified it by reading level with machine learning, and delivered it to students through APIs, SDKs and classroom apps.

2014–2021 · CTO
AWSMachine learningElasticsearchJavaReactCI/CD

The company

Wizenoze is an ed-tech scale-up based in Amsterdam. It set out to give students a safer and more educational way to find things than general search engines and content services: material matched to their curriculum and pitched at their reading level.

I was its CTO from July 2014 to April 2021, working from Cape Town and Amsterdam, and ran the full product development cycle.

The role

  • Recruiting, managing, leading and mentoring the development team.
  • Architecture and design guidance for the team, while still building full-stack services myself in Java, Groovy and Spring, with React/Redux and RiotJS frontends.
  • API design, and the distributed systems on AWS.
  • CI/CD automation and DevOps.
  • Buy, build or partner decisions on technology.
  • Product development from the users all the way down to the hardware.
  • Technology and business strategy.

What we built

  • A web crawler, first on Apache Nutch and then on StormCrawler, collecting content from the open web.
  • A service-oriented platform of APIs that carried every frontend and product, and the APIs customers integrated with.
  • Web for Classrooms, a search engine over human-curated websites, crawled and classified with machine learning. Elasticsearch and custom-tuned ranking brought back the most relevant educational content for students.
  • Wizeup, which matched content to mapped curricula across primary, secondary and vocational education, and delivered it through APIs, SDKs and standalone web apps at the right place and the right time.
  • WizeNote, a content editor that used machine learning and NLP to guide writers in adapting text to lower reading levels.
  • The delivery pipeline: full CI/CD on Jenkins, Docker and Elastic Beanstalk, with Sentry for errors and Prometheus and Grafana for monitoring and dashboards.
  • Data analysis pipelines in Kibana and Klipfolio.
  • A run of MVPs along the way.

Need a CTO in the room?

Tell me what is going on. The first conversation is free, and I will say honestly whether I can help. If I can’t, I’ll introduce you to someone who can.

Cape Town, working remotely. My mornings overlap with Australian afternoons, and my afternoons with the UK, Europe and the US east coast. Or message me on LinkedIn.