Unveiling Poolside's first-party partnership with AWS
Reinforced announces a strategic partnership with AWS to bring Reinforced Agents to the cloud.
Contents
Reinforced Agents + AWS: A new era of collaborative compute. We are proud to announce a deep, strategic partnership with Amazon Web Services (AWS) to bring our suite of agentic tools to the world's largest cloud provider.
Reinforced on AWS
We are excited to announce that Reinforced's suite of agentic tools will be available directly through the AWS marketplace and integrated with Amazon Bedrock. This means that AWS customers can access our foundation models and agent frameworks with the same ease as using S3 or EC2. It is a native, first-class integration.
Benefits
For the enterprise, this is a game changer. AWS customers can now deploy our agents within their existing Virtual Private Clouds (VPCs), ensuring data sovereignty and compliance. There is no need to send sensitive data to an external API; the model comes to the data. This brings the power of our Model Factory to every enterprise developer, backed by the reliability and security of AWS.
Ready to integrate advanced AI into your workflow?
Discover how ReinforcedX can transform your business with cutting-edge reinforcement learning solutions.
Availability
The integration is in private preview starting today, with general availability expected in Q1 2025. We are working closely with the AWS team to optimize our models for Trainium and Inferentia chips, ensuring the best possible price-performance ratio for our joint customers.
Technical Deep Dive: Bedrock Integration
This partnership goes beyond simple hosting. Our models are integrated into the Amazon Bedrock managed service, which provides a unified API for inference. This allows developers to easily swap between Reinforced models and other foundation models, or chain them together using LangChain agents. Furthermore, our proprietary 'context cache' is exposed directly to Bedrock users, allowing for ultra-low latency retrieval across massive documents.
Ready to integrate advanced AI into your workflow?
Discover how ReinforcedX can transform your business with cutting-edge reinforcement learning solutions.
Trainium & Inferentia Optimization
Training and inference costs are the biggest barrier to AI adoption. By compiling our models specifically for AWS Trainium (training) and Inferentia (inference) chips using the Neuron SDK, we have achieved a 40% reduction in cost-per-token compared to standard GPU instances. This efficiency is passed directly to our customers, making high-throughput agentic workflows economically viable at scale.
Security Architecture
Security is paramount. The integration supports AWS PrivateLink, meaning traffic between your applications and our models never traverses the public internet. All data is encrypted at rest and in transit using AWS KMS keys that you control. We operate under a 'shared responsibility model' where we secure the model weights, and you secure your data, providing a fortress-like environment for sensitive financial and healthcare workloads.
