-
Mastering the Future of Autonomous Driving with End-to-End AI
Explore how end-to-end AI models are reshaping autonomous driving by replacing modular perception-planning-control pipelines with unified neural networks. Discusses vision-language models, sensor fusion strategies, and the infrastructure required to train and validate these systems at scale on AWS.
-
How to Deploy Trained Models to Robots After Using Functions Generated by Large Models
Bridge the gap between foundation model outputs and real-world robotic deployment. Covers the workflow of using large language models to generate task functions, training robot policies via simulation, and deploying optimized models to edge devices for reliable physical execution.
-
How to Fine-Tune a 6B LLM Simply and Cost-Effective Using Ray AI Runtime on Amazon EKS
Fine-tune a 6-billion parameter language model using Ray AI Runtime on Amazon EKS without managing complex distributed training infrastructure. Demonstrates parameter-efficient tuning with LoRA, Ray Train for distributed scaling, and cost optimization strategies using spot instances and right-sized GPU nodes.
-
How to Build a Successful Open Source Project for Autonomous Driving
Lessons learned from building and growing open source projects in the autonomous driving space. Covers community engagement strategies, contribution guidelines, modular architecture design, and how to balance enterprise requirements with open collaboration to accelerate industry-wide innovation.
-
Augmenting Datasets using Generative AI and Amazon SageMaker for Autonomous Driving Use Cases on AWS
Use Generative AI and Amazon SageMaker to synthetically augment autonomous driving datasets. Generate diverse weather conditions, lighting variations, and road scenarios to expand training data coverage and improve ML model robustness without additional real-world data collection.