Senior AI Agent Developer – Gemini ADK & Enterprise

TechDigital Group; Secaucus (NJ); On-site; USD 120,000 - 160,000; Full time

Job summary

TechDigital Group in the United States is seeking a Senior AI Engineer to design and build enterprise-grade AI agents on Google Gemini Enterprise using the ADK for Finance, Marketing, and HR use cases. You will implement agent logic, memory, tools, prompts, and guardrails.

Work closely with Data & AI Engineers and MCP Developers across sprints; contribute reusable components to Client Agent Library, and optimize prompts and model configuration.

Qualifications

• Hands-on Google ADK and Gemini Enterprise agent development.
• Strong Python and prompt engineering for LLM/GenAI applications.
• Experience with RAG, embeddings, vector search, and grounding.
• Understanding of agent orchestration (A2A) and MCP tool integration.

Responsibilities

• Develop AI agents using Google ADK, Gemini foundation models, and the A2A protocol.
• Implement agent reasoning, planning, tool invocation, memory, and sub-agent orchestration.
• Integrate agents with enterprise data via MCP and RAG using Databricks, ServiceNow, Snowflake, OneDrive, Adobe, PDFs.
• Build reusable, modular agent components and publish to the Client Agent Library.
• Perform unit and evaluation testing using the ADK eval framework; optimize prompts and model configuration.
• Collaborate with Data & AI Engineers, API/MCP Developers, and FDEs across the sprint lifecycle.

Skills

Google ADK

Gemini Enterprise

Python

Prompt engineering

RAG

A2A orchestration

MCP integration

Tools

Databricks

ServiceNow

Snowflake

OneDrive

Adobe

PDF handling

Job description

TechDigital Group in the United States is seeking a Senior AI Engineer to design and build enterprise-grade AI agents on Google Gemini Enterprise using the ADK for Finance, Marketing, and HR use cases. You will implement agent logic, memory, tools, prompts, and guardrails.

Work closely with Data & AI Engineers and MCP Developers across sprints; contribute reusable components to Client Agent Library, and optimize prompts and model configuration.