backend developer for AI agents

EPAM develops enterprise software products, open source solutions, and accelerators. The role contributes to an Agent Gateway for a cloud-native Enterprise Agent Development Platform that enables engineering teams to define, orchestrate, deploy, and observe AI agents at scale.

• Develop and maintain A2A Gateway components for agent discovery, registration, and interaction;
• Implement the Agent Card JSON schema and /.well-known/agent.json endpoints for identity and configuration publishing;
• Design and integrate A2A invocation patterns using JSON-RPC 2.0 messaging, task lifecycle events, and push notifications;
• Build and optimize high-performance APIs in Python with async frameworks and AgentCore SDK;
• Implement secure and efficient communication flows for agent interactions using REST and SSE streaming;
• Collaborate with platform and orchestration teams to validate integration correctness and performance under load;
• Ensure observability of inter-agent workflows, supporting traceability and debugging;
• Support deployment topologies for specific gateway use cases and configurations.

• 3+ Years of experience as a backend engineer working with Python or Go;
• Strong experience building and integrating APIs, including routing and security considerations;
• Hands-on experience implementing REST APIs and asynchronous/event-driven patterns such as SSE and webhooks;
• Understanding of microservice communication and integration patterns in production environments;
• Familiarity with schema design and structured JSON endpoints for distributed services;
• Nice to have: Experience with AWS AgentCore Gateway and related templates, exposure to AWS Bedrock multi-agent invocation patterns, knowledge of SSE event streaming for task lifecycle handling and scalability, familiarity with OpenTelemetry for distributed tracing propagation across A2A agent flows.

• Remote work in Portugal.