Notice: According to its own product page, AWS now provides DeepRacer as an open-source solution: customers deploy and operate the environment in their own AWS account (implementation guide “DeepRacer on AWS”, published January 2026, with a CloudFormation template and source code on GitHub). AWS has not published an official shutdown date for the previous managed service. Training still runs on Amazon SageMaker AI.
What is AWS DeepRacer?
AWS DeepRacer is an educational tool for learning reinforcement learning (RL). The service combines a 1:18 scale autonomous model car with a cloud-based 3D racing simulator and a global racing league. Developers train RL models, define reward functions, and optimize their agents to achieve the fastest lap time.
DeepRacer makes reinforcement learning tangible: instead of abstract environments, the trained agent controls a real or simulated racing vehicle. The reward function defines what behavior is desired (staying on track, driving fast), and the model learns the optimal strategy through thousands of training rounds.
Core Features
- 3D Simulator: Realistic racing environment for training and testing without a physical car
- Reward Function Editor: Python-based editor for custom reward functions
- Model Training: Cloud-based RL training with SageMaker integration
- Racing League: Global competitions with virtual and physical races
- Multi-Car Racing: Train models for racing against other agents
Typical Use Cases
RL Education: Development teams learn reinforcement learning through hands-on experimentation. The racing format makes the learning process motivating and results immediately visible.
Team Building and Hackathons: Companies organize internal DeepRacer competitions as team building events. Teams compete with their RL models against each other and learn ML fundamentals in a competitive format.
ML Strategy Evaluation: Developers experiment with different RL algorithms (PPO, SAC) and reward functions, developing an understanding of the challenges of model training.
Advantages
- Makes reinforcement learning accessible through gamification
- Free simulator without hardware investment
- Global community and competitions for motivation
- Direct application of SageMaker skills
Integration with innFactory
As an AWS Reseller, innFactory supports you with AWS DeepRacer: organization of RL workshops, hackathons with DeepRacer competitions, and transition from educational RL projects to productive ML applications.
Typical Use Cases
Frequently Asked Questions
What is AWS DeepRacer?
AWS DeepRacer is a 1:18 scale racing car powered by reinforcement learning (RL). The service includes a physical car, a 3D racing simulator, and a global racing league. Developers train RL models in the cloud and test them virtually or physically.
What is reinforcement learning?
Reinforcement learning is an ML technique where an agent learns through trial and error. The agent (the car) performs actions (steering, accelerating), receives rewards or penalties, and optimizes its strategy over thousands of training rounds. DeepRacer makes this concept tangible.
Do I need the physical car?
No, the 3D simulator works entirely without hardware. You train and test RL models in the virtual racing environment and participate in virtual races. The physical car additionally enables transferring models to the real world.
Note: All product information on this page has been compiled with care, but is provided without guarantee and may be outdated or incomplete. Cloud services evolve rapidly — features, pricing, SLAs, and availability change frequently. Authoritative and up-to-date information can only be found on the official product page of AWS (official documentation). This page does not represent an offer by AWS.