The challenges and frontiers of data science in Europe
summary
Panel discussion reports concrete wins and failures: a hedge‑fund manager wires up an numpy convolution experiment to test a new signal, Royal Mail commits to hire a ten‑person data team and a budget, and a Google‑style search‑term flu predictor shows direction for productized analytics. Persistent blockers are messy schemas and logging, so teams adopt protocol buffers for schemas and argue for data in the SDLC. Tool takeaways: pandas numpy scikit‑learn Anaconda and deployment tradeoffs are debated in detail.
AI-written summary · about
chapters (10)
- 0:00Data science in Europe intro
- 15:59Value of switching to data science
- 23:13Convincing managers to invest
- 33:45Expert systems to data-driven
- 39:22Industries missing in Europe
- 46:20Python 2 vs Python 3 debate
- 48:51US vs European data science
- 53:08Missing Python skills
- 57:25Cluster computing needs
- 58:34Audience enthusiasm for Python
related talks
27mData, Python and Humans
01h 16mLightning Talks (01)
48mBecoming a Data Head, a reflection on the trade offs and decision
45mPyData London 2018 Lightning Talks - Day 2
34mData Science beasts (failures) and where to find them
25mBuilding A Successful Data Science Team
36mPyData London 2017, Lightning Talks - Saturday
03mPyData Amsterdam 2018