Описание
Intellias is building an enterprise-grade AI Agent Platform that provides a standardized foundation for developing, deploying, orchestrating, securing, and observing AI agents across multiple teams and use cases. The project brings together agent orchestration, observability, security, governance, integrations, evaluation, and platform engineering.
Задачи
- Design and implement MCP server and client components in Python.
- Implement MCP capabilities, including Streamable HTTP transport, tool discovery and invocation, and capability negotiation.
- Build asynchronous HTTP services and support streaming communication patterns.
- Develop integrations that expose AWS Lambda functions, remote MCP servers, and APIs as agent tools.
- Implement OpenAPI-to-MCP translation and standardized tool registration mechanisms.
- Define and manage tool schemas, parameters, metadata, capability tags, and versioning.
- Integrate tools with LLM agents using function calling and agent tool integration patterns.
- Build parameter validation and error-handling mechanisms for tool execution.
- Develop and maintain AWS Lambda-based tool integrations.
- Integrate MCP tools with agent frameworks such as LangChain, LangGraph, or Strands.
- Contribute to semantic tool discovery and indexing, including embeddings for tool descriptions.
Требования
- 4+ Years of Python engineering experience with agent/AI systems.
- Experience building tool integrations for LLM agents, such as OpenAI functions, LangChain tools, or equivalent.
- Experience implementing async HTTP services and streaming protocols.
- AWS Lambda development experience.
- Nice to have: MCP protocol implementation experience (server or client), LangChain / LangGraph tool integration, AWS Bedrock AgentCore Gateway target registration, semantic tool indexing (embedding generation for tool descriptions), agent framework integration (Strands, Crew, AutoGPT).
Условия
No conditions specified