Machine learning,
applied
honestly
jynovue is a Dublin-based workshop platform built around one specific problem: most machine learning courses teach theory in isolation, without connecting it to the messy, real data that practitioners actually encounter. We focus on cryptocurrency market data because it is public, fast-moving, and genuinely difficult to model - which makes it an effective training ground for building analytical skills that transfer elsewhere.
Where the platform came from
The idea behind jynovue grew out of a frustration with how machine learning education was structured. Courses would walk through clean, pre-processed datasets and leave participants without the skills to handle noisy, inconsistent real-world data. Cryptocurrency price and volume data - available freely via exchange APIs - turned out to be an ideal subject: it is genuinely complex, well-documented, and relevant to a wide range of analytical questions.
The workshops are designed around assignments that require participants to make decisions, not just follow steps. Each session introduces a concept through a concrete scenario - for example, building a feature set from OHLCV data, or evaluating whether a trained model is actually learning signal or fitting noise. The goal is to develop judgment, not just technique.
The hardest part of applied ML is not the maths - it is deciding what the data is actually telling you, and when to trust it.
Workshop access options
Each tier gives access to the same core curriculum. The difference is in how much direct feedback and collaboration you receive during the process.
- Full workshop materials
- Step-by-step assignments
- Dataset access for exercises
- Community forum participation
- Everything in Starter
- Weekly live review sessions
- Assignment feedback from instructor
- Peer group of 8–12 participants
- Extended dataset library
- Everything in Standard
- 1-to-1 mentoring sessions
- Custom project scope
- Priority instructor access
- Certificate of completion
Workshops run in structured cohorts with fixed start dates. Each session builds on the previous one, so participants develop a working understanding of the full pipeline - from raw data ingestion through to model evaluation - rather than isolated fragments of knowledge.
