ML Deployment Methodology Illustrated by Route Time Estimation
summary
The talk covers applying machine learning to predict express delivery travel times, describing the problem of delivery time estimation and how ML models are integrated into routing algorithms. It presents a target system architecture for this ML-enabled forecasting and highlights practical considerations and important issues to address during deployment.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 2:13What is Yandex Delivery
- 7:09Target task and architecture
- 13:17Tech stack and solution approach
- 17:37ML in production challenges
- 22:36Focus on results and first steps
- 26:51Faster time to production
- 30:47Servicing time component implementation
- 34:46Optimization and speed balance
- 36:14Results and deployment outcomes
- 38:24Conclusions and recommendations
- 48:24Q&A
related talks
43mML application lifecycle
12mUse ML to Predict Delivery Time and Improve Code Reliability
44mML in Production
39mMethodology of ML in Performance Testing (How to Find Bottlenecks Automatically)
18mExtending Mobility - How AI Contributes towards the Future of Railway Transportation
30mChallenges of machine learning: The new ML lifecycle
47mMLOps tools to automate ML deployment
17mContinuous deployment for ML: The new software development lifecycle