On the Delivery of Data Science Projects

Talk at PyDataCambridge 2019-05 on the two halves of Data Science delivery - derisking the business side of the project and improving the software engineering side of the project. Contains observations, issues, new tools and process ideas you can use back in your teams.

Transcript

1. On the Delivery of Data Science
   Projects
   @IanOzsvald – ianozsvald.com
   Ian Ozsvald
   PyDataCambridge 2019-05
   

   View Slide

2. 
   Interim Chief Data Scientist
   
   19+ years experience
   
   Quickly build strategic data science plans
   
   Team coaching & public courses
   Introductions
   By [ian]@ianozsvald[.com] Ian Ozsvald
   

   View Slide

3. 
   Numerate management ask good data-driven questions
   
   You have suitable data
   
   Well defined achievable outcomes are defined
   
   Change is enabled by these projects
   Data Science shows value when...
   By [ian]@ianozsvald[.com] Ian Ozsvald
   

   View Slide

4. 
   Unclear true business need
   
   No visibility on the data (and its quality)
   
   Blind belief in 100% success
   
   No project specification – lacking shared agreement
   Common failure points
   By [ian]@ianozsvald[.com] Ian Ozsvald
   

   View Slide

5. 
   What’s the driver? Is there a fire under it?
   
   Joonatan’s example from PyDataLT – OCR
   
   Cost/benefit estimate accepting uncertainty
   Checking business need
   By [ian]@ianozsvald[.com] Ian Ozsvald
   

   View Slide

6. 
   States a clearly defined problem
   
   Guesses at unknowns (and project torpedoes!)
   
   Proposed milestones and Gold Standard/metrics
   
   Clear “definition of done”
   
   Story from 10 years back
   You need a Project Specification
   By [ian]@ianozsvald[.com] Ian Ozsvald
   

   View Slide

7. 
   Do you understand your data?
   – What’s good and bad?
   – What relationships exist?
   
   Build exportable Notebook as html artefact
   
   Read Bertil’s piece on Medium
   “Data Story”
   By [ian]@ianozsvald[.com] Ian Ozsvald
   

   View Slide

8. 
   Easy first deliveries – reports
   
   Get to a minimal working delivery as soon as
   possible
   
   Two tracks? R&D and client integration?
   Continuous delivery to clients
   By [ian]@ianozsvald[.com] Ian Ozsvald
   

   View Slide

9. 
   Reduce mental load for common decisions
   – Cookiecutter data-science
   – Watermark
   – Pandas-profiling / edaviz
   – Anaconda
   Standardised Approaches
   By [ian]@ianozsvald[.com] Ian Ozsvald
   

   View Slide

10. Code quality
    By [ian]@ianozsvald[.com] Ian Ozsvald
    Attrib: https://devrant.com/rants/347670/code-quality-as-measured-in-wtfs-minute
    

    View Slide

11. 
    Encode assumptions using asserts (example -
    yesterday’s client issue)
    
    Refactor to modules
    
    Add unit-tests
    
    Diagnostics e.g. yellowbrick for sklearn
    Continuously improving code quality
    By [ian]@ianozsvald[.com] Ian Ozsvald
    

    View Slide

12. 
    Exposure to new processes
    
    Enforced clear communication
    
    Balanced consumption & contribution
    
    You’re more visible & valuable
    Contributing to Open Source gets you
    By [ian]@ianozsvald[.com] Ian Ozsvald
    

    View Slide

13. 
    High test coverage
    
    Easy roll out & roll back
    
    Culture of constructive criticism
    High performance teams
    By [ian]@ianozsvald[.com] Ian Ozsvald
    

    View Slide

14. “Successfully Delivering
    Data Science Projects” &
    “Software Engineering for Data
    Scientists” - early July
    Resources
    By [ian]@ianozsvald[.com] Ian Ozsvald
    

    View Slide

15. 
    Derisk early and often
    
    Communicate visually, all the time
    
    Strive to continuous improvement
    
    Join my thoughts+jobs list for tips and my training list
    
    Attend PyDataLondon 2019 July 12-14?
    Summary
    By [ian]@ianozsvald[.com] Ian Ozsvald
    

    View Slide