Note: The job is a remote job and is open to candidates in USA. Nasscomm is seeking a highly skilled GCP AI Engineer to lead the design, development, and deployment of enterprise-grade, multi-agent orchestration systems. The role involves leveraging the Gemini Enterprise Agent Platform to build autonomous AI workflows and ensuring that all AI deployments adhere to strict enterprise security, governance, and compliance guardrails.
Responsibilities
- Multi-Agent Orchestration & Development
- Architect Multi-Agent Systems: Design and implement multi-agent workflows using the Gemini Enterprise Agent Platform, Google Agent Development Kit (ADK), LangGraph, and/or CrewAI
- Complex Workflow Automation: Develop long-running agents capable of managing multi-day, multi-step workflows, utilizing complex reasoning patterns (e.g., ReAct, hierarchical delegation, self-reflection)
- Tool & System Integration: Architect the "connective tissue" between our AI agents and live enterprise infrastructure (e.g., Oracle database Google Workspace, Jira, Salesforce, BigQuery) using Model Context Protocol (MCP) servers and custom APIs and Oracle Database MCP server
- Agentic RAG: Build and optimize multi-agent Retrieval-Augmented Generation (RAG) pipelines that include Planner Agents, Query Rewriters, and Orchestrators to iteratively search and synthesize data
- Security, Governance & Compliance
- Agent Governance: Utilize Agent Registry as the single source of truth to index, manage, and discover approved internal agents, tools, and skills
- Network & Access Security: Deploy and configure Agent Gateway as the unified network control point. Implement Agent Identity (with end-to-end mTLS) and Identity-Aware Proxy (IAP) to validate agent permissions against fine-grained IAM policies before allowing tool access
- AI Guardrails: Integrate Model Armor and Service Extensions to screen agent prompts and responses, protecting against prompt injection attacks, hallucinations, and data leakage
- Safe Execution Environments: Ensure secure code execution and browser-based automation by utilizing Agent Sandbox isolated environments
- MLOps & Performance Optimization
- Deployment & CI/CD: Build production-grade ML pipelines for deploying, monitoring, and updating agentic workflows on Google Cloud
- Observability & Evaluation: Build high-performance evaluation pipelines using "LLM-native" metrics (e.g., tokens/sec, cost-per-request). Optimize state management, context passing, and granular tracing across agent interactions
Skills
- Bachelor's degree in Computer Science, Engineering, AI/ML, or a related field
- 7+ years of software engineering experience using Python, Go, or similar languages
- Deep hands-on experience with Google Cloud Platform (GCP) and deploying applications via Cloud Run, GKE, or similar compute environments
- Proven experience building and deploying Large Language Models (LLMs) and Generative AI solutions using Gemini, Vertex AI, or similar frontier models
- Experience building structured and unstructured data pipelines, Vector Databases, and production-grade RAG architectures
- Solid understanding of cloud security principles, IAM, VPCs, and secure API integrations
- Master's degree or PhD in AI, Computer Science, or a related technical field
- Hands-on experience with the Gemini Enterprise Agent Platform and Google's Agent Development Kit (ADK)
- Proven track record of implementing multi-agent systems using modern frameworks (LangGraph, AutoGen, CrewAI)
- Expertise in AI governance and security tools (e.g., Model Armor, Agent Gateway, Data Loss Prevention for AI)
Company Overview