Building a Machine Learning Platform at

on YouTube ↗18m

" summary

Quora describes a reusable, extensible machine-learning platform used to productionize dozens of models for tasks like content quality assessment and user interest/expertise detection. The talk covers the platform’s core ideas and the specific systems, tools, and abstractions their ML engineering team uses to scale and deploy models across Quora.

AI-written summary · about

" chapters (7)

1. 0:00Start
2. 6:55Fast Scalable Production Systems
3. 8:37Blurry Line Between Experimentation & Production
4. 9:22Openly Using Open Source
5. 10:12SPARKLING WATER
6. 12:59Blurry Line Between ML & Product Dev
7. 17:11Summary

" related talks

• 03mInterviewJulie Pitt · Scale By The Bay 2018
• 32mTeam Teardown: Evolving an ML PlatformSam Charrington, Jing Huang, Shubhi Jain, Jolene Zhou · TWIMLcon: AI Platforms 2019
• 42mXavier Amatriain, VP of Engineering, QuoraXavier Amatriain · MLconf SF 2015
• 25mXavier Amatriain, VP of Engineering, Quora @ MLconf SEAXavier Amatriain · MLconf Seattle 2015
• 38mFrom text to knowledge via ML algorithms - the Quora answerXavier Amatriain · Data By the Bay 2016
• 27mML Platform JourneyJessica Egoyibo Mong · TWIMLcon: AI Platforms 2019
• 50mDemocratizing AI - Back-fitting end-to-end machine learningJoel Young, Bo Long · Strange Loop 2018
• 29mBuilding ML Platform Capabilities at Global ScalePaul van der Boor · TWIMLcon: AI Platforms 2021