True ML Talks #9 | Machine Learning @DoorDash
True ML Talks #9 | Machine Learning @DOORDASHINGCMOTO
In today's episode of #TrueMLTalk, we are joined by Hien Luu, the senior engineering manager at DoorDash. Hien and his team have been working on building a centralized ML platform to enhance the development of machine learning projects at DoorDash. Over the years, they have managed to rapidly accelerate their ML journey and achieve significant progress in a short span of time.
Join us as we dive into the details of their journey, exploring the architecture of their ML platform, the impact it has had on their business, and the lessons learned along the way.
In our conversation with Hien, we covered the following aspects:
✅What prompted DoorDash to start using ML?
✅What is the scale of the ML platform at DoorDash
✅The fundamental principle that led to building their ML platform
✅The impact within the business side that led them to investing more in ML
✅The architecture behind the model-serving layer at DoorDash
✅The reason behind choosing GRPC as a standard protocol
✅How the ML shadowing layer was built and its impact in terms of deployment velocity
✅What were the parts where the data scientist was enabled to do things freely?
✅What does the training workflow look like for the data scientists?
✅How are the different data formats handled at DoorDash
✅Challenges faced in terms of cost, resources, and performance while making the platform
✅The use cases for generative AI being explored at DoorDash
✅The importance of hosting LLMs internally for privacy reasons
Liked the video? Like, comment, and subscribe for more!
_____________________________________
00:00 Start.
02:40 The journey of ML at DoorDash and its business impact
04:42 The scale of DoorDash’s ML platform scale
07:58 The fundamental principles that were considered when building the platform
12:45 The impact of ML on the logistics team and last-mile delivery
15:12 The architecture of model-serving layer & how it supports billions of predictions daily
18:30 Overview of the shadowing and canary model
21:12 How the shadowing layer was built
24:48 How does the platform enable data scientists to work freely?
27:50 An overview of the entire training pipeline and the data warehousing layer
35:01 How does DoorDash handle failed model training and debugging
39:17 How DoorDash uses model monitoring to measure drifts
41:20 Challenges faced while building the ML platform at DoorDash
44:15 Impact of Gen-AI and LLM on DoorDash's ML
___________________________________
ABOUT OUR GUEST
Hien Luu is a highly experienced senior engineering manager at DoorDash who has been instrumental in building a centralized ML platform to improve the development process at the company. He is passionate about big data and machine learning, and as a part of DoorDash, his role is to oversee various ML use cases across the company's product lines, ensuring that they are aligned with the company's overall strategy and business goals, among other duties.
(This is all the information that is provided on the transcript due to poor audio quality)
⚡Reach out to Hien at the below link - / hienluu
ABOUT OUR CHANNEL
TrueMLTalks is a video series in which we interview machine learning industry professionals from companies such as Gong, StichFix, SalesForce, Facebook, Simpl, and others. We provide an insightful understanding of their experiences managing complex ML pipelines and developing successful best practices, making it a valuable resource for professionals looking to stay current on the latest advances in the field.
_____________________________________
ABOUT TRUEFOUNDRY
TrueFoundry is a cross-cloud machine learning deployment PaaS that enables enterprises to speed up model testing and deployment while maintaining full security and control for the Infra/DevSecOps team. We enable machine learning teams to deploy and monitor models in 15 minutes with 100% reliability and scalability, saving money and allowing models to be released into production faster, resulting in genuine business value. We deploy on the customer's infrastructure, taking data privacy and other security concerns into consideration.
⚡ A short ~4 min demo of the platform: https://t.ly/0fk1
Read our blog: https://blog.truefoundry.com/
Subscribe to our newsletter: https://t.ly/oORp
Try our platform now - https://app.truefoundry.com/company-r...
Book a personalised demo - https://www.truefoundry.com/book-demo...
#TrueFoundry #machinelearning #mlops #datascience
Watch video True ML Talks #9 | Machine Learning @DoorDash online without registration, duration hours minute second in high quality. This video was added by user TrueFoundry 11 May 2023, don't forget to share it with your friends and acquaintances, it has been viewed on our site 484 once and liked it 24 people.