Agentic AI Engineer For Remote Role

Design and implement a "super-agent" orchestrator using Vertex AI Agent Builder, Google's Agent Development Kit (ADK), the Model Context Protocol (MCP) and Agent-to-Agent (A2A) protocol. You'll own end-to-end delivery-from defining multi-agent workflows and writing Python code, to deploying agents, securing MCP tool endpoints, and enabling a Teams chatbot channel.

Key Responsibilities

Architect & Implement Orchestrator

Using ADK's SequentialAgent, ParallelAgent, or custom workflow primitives, compose and code a master orchestrator that invokes, monitors, and aggregates multiple specialized sub-agents.

Register and configure MCP-compliant "tools" (APIs, database queries, embedding services) and enable A2A message-passing for dynamic delegation.

Build & Deploy Agents

Develop individual LLM agents (e.g. retrieval, classification, summarization) in Python, each with clear instructions, tooling interfaces, and guardrails.

Containerize agents and deploy via Vertex AI Agent Builder / Reasoning Engine; configure autoscaling, logging, and CI/CD pipelines.

Integrate with Microsoft Teams

Set up Agentspace's Teams connector to ingest chat context; configure the Copilot Studio channel so end users can interact with the orchestrator inside Teams.

Ensure secure OAuth/mTLS flows for both data ingestion and agent invocation.

Security, Observability & Testing

Implement authentication, authorization, and audit logging around all MCP tool calls.

Instrument distributed tracing and monitoring for A2A handoffs; write integration tests for multi-agent workflows.

Documentation & Knowledge Transfer

Write clear design docs, API specs, and runbooks.

Mentor in-house team members on Agentic workflows best practices.

Required Qualifications
10+ Years of Overall experience

6+ years hands-on experience in Python software engineering, with a focus on AI/ML or large-scale microservices

3+ years on Google Cloud Platform (Vertex AI, Cloud Run, Cloud Workflows, IAM)

Familiarity with:

Vertex AI Agent Builder & Agent Development Kit (ADK)

Model Context Protocol (MCP) and Agent-to-Agent (A2A) protocol concepts

Containerization (Docker) and Kubernetes or Cloud Run

Experience integrating applications with Microsoft Teams (Graph API, Teams connectors, Copilot Studio)

Strong API design skills (REST/gRPC), OAuth2.0, mTLS, and secure microservices patterns

Proven track record of architecting, deploying, and debugging distributed workflows

Excellent communication, documentation, and collaboration skills

Preferred / Nice-to-Have

Prior work with LangChain, LlamaIndex or similar agent orchestration frameworks

Background in conversational AI, RAG pipelines, or embedding/search architectures

Familiarity with CI/CD tooling (Cloud Build, GitHub Actions, Terraform)

Knowledge of observability stacks (OpenTelemetry, Stackdriver Logging/Tracing)

Experience in agile teams and mentoring juniors