AFM 201705

Transcript

1. SUCCESSFUL DATA SCIENCE Giovanni Lanzani – Chief Science Officer GoDataDriven gdd.li/afm
2. WHO AM I • Imported from Italy ca. 2006 • Master & PhD in Theoretical Physics in Leiden (2006-2012) • Consultant Software Quality @ KPMG (2012-2013) • Data Whisperer @ GoDataDriven (2013-2016) • Chief Science Officer @ GoDataDriven (2016-…)
3. ABOUT GODATADRIVEN • Founded in 2013, part of the Xebia group (600 FTE) • 30 machine learning engineers, data scientists, data engineers • Commercial focus on Dutch top 200 or category leaders • Based in Amsterdam
4. CLIENTS
5. WHAT IS MACHINE LEARNING
6. LEARNING FROM DATA • You have some (lots) of data • You need to generalize
7. BEST MODEL • Which one would you choose here? • It’s about making a tradeoff • This trade off is the most important job of the PO • A 100% correct answer might not exist!!!
8. WHAT’S DATA SCIENCE
9. ULTIMATELY • It’s about creating value from data • Using Machine Learning, Advanced Analytics, and visualization
10. HOW TO GET THERE Beyond the data warehouse
11. BEYOND THE DATA WAREHOUSE Traditional operational data sources EDW Data consumer Web app Dashboard / Reporting Traditional Business app
12. DATA PLATFORM Machine Learning Data pipelines Scale Data consumer Web app Dashboard / Reporting Traditional Business app API External API Logs Chat/transcripts Scraping Unstructured data Traditional operational data sources
13. PROCESS
14. PROCESS Requirements Data Sources Exploration Modeling Products Feedback
15. PEOPLE Requirements Data Sources Exploration Modeling Products Feedback Data scientist ML engineer Data engineer Data engineer
16. SKILLS
17. JOB MARKET 2016 • The biggest differentiator I've seen is to be able to participate in actually building production quality systems vs being proficient enough in R or python to hack together a prototype on a very small dataset • Supply of the second group keeps growing while demand is flat or shrinking • especially as executives get burned by “data scientists” who don't know how to help them build things of value
18. JOB MARKET 2017 • I am seeing the same things happening • We (GoDataDriven) are definitely only interested in these profiles (people who are already there, or that are getting there) • Many of our clients are in the same position
19. HIRING • Companies that are not engineering driven, often have trouble hiring good technical people • The “IQ” test is not really representative of applied data science • At GoDataDriven we do a “at home, at your convenience” assessment • Real dataset, real business question, real product
20. KAGGLE CURSE • Many data scientists approach the problem at hand with a Kaggle-like mentality: delivering the best model in absolute terms, no matter what the practical implications are. • In reality it's not the best model that we implement, but the one that combines quality and practicality. • Netflix competition
21. ORG STRUCTURE One Team to rule them all, One Team to find them, One Team to bring them all and in the data bind them
22. CENTRAL TEAM • Highly skilled • They only do the difficult parts • Easier to attract people with everchanging challenges • Easier to cross pollinate knowledge • Formalized and efficient process • Backlog out of the business • Prioritization is unclear • Maintenance and continuous improvement is done by different people
23. MULTIPLE TEAMS • Backlog in the hand of the business • Maintenance and improvement are (mostly) done by the same people • More difficult to have many good teams • Risk of reinventing the wheel everywhere • Sharing knowledge is not as easy • Lots of different standards • Difficult professionalization as the business has different priorities
24. QUESTIONS?
25. BUILDING A DATA PRODUCT
26. Data Platform DATA CONTRACTS BU BU BU Data Lake Data Products Business Business Business IT IT IT App App App external data collection internal data collection
27. BUILDING A DATA PRODUCT • Data {insert something here} should be pro grade