GenAI Lead / GCP AI Engineer Lead

Capgemini is building production GenAI capabilities for insurance clients on Google Cloud, with a focus on taking architecture from design through working software in production. In this role, you will build, test, and deploy GenAI solutions that combine Gemini models, Vertex AI agent services, and GCP data and document tooling to deliver assistants and agentic workflows.

What you’ll do

• Develop GenAI applications using Gemini models on the Gemini Enterprise Agent Platform (Vertex AI).
• Design and tune prompts, structured outputs, and function calling for insurance document workflows involving submissions, ACORD forms, COIs, loss runs, and policies.
• Build RAG pipelines including chunking, embeddings, Vector Search / Vertex AI Search, grounding, and citations.
• Create agents with the Agent Development Kit (ADK) and deploy them on Agent Engine, integrating tools and APIs.
• Extract and classify documents using Document AI with Gemini multimodal input.
• Ship services on Cloud Run or GKE, expose them through Apigee, and implement CI/CD via Cloud Build.
• Move and store data with BigQuery, Cloud Storage, Pub/Sub, and Dataflow.
• Create evaluation sets and run quality, latency, and cost checks using a Gen AI evaluation service and golden datasets.
• Apply guardrails including PII masking with Sensitive Data Protection (DLP), safety filters, and IAM least privilege.
• Collaborate in Agile sprints with QA, UI, and data teams, writing clean, tested, and documented code.

What you bring

• Strong Python skills (for example FastAPI or Flask), including async development and pytest.
• Hands-on experience with Gemini API / Vertex AI SDK or the Google Gen AI SDK.
• Practical RAG experience and experience with at least one vector store such as Vector Search, AlloyDB pgvector, or BigQuery vector search.
• Experience with LLM frameworks including ADK, LangChain, or LangGraph.
• Comfort with core GCP services including Cloud Run, Cloud Functions, Cloud Storage, BigQuery, IAM, and Secret Manager.
• Experience with REST API design, Git, Docker, and CI/CD.
• Good written and verbal communication with onsite and client teams.

Tools and technologies you may use

• Python, FastAPI, Flask, pytest
• Gemini API, Vertex AI SDK, Google Gen AI SDK
• Gemini Enterprise Agent Platform (Vertex AI), Agent Development Kit (ADK), Agent Engine
• Document AI, Gemini multimodal input
• Cloud Run, GKE, Apigee, Cloud Build
• BigQuery, Cloud Storage, Pub/Sub, Dataflow
• Sensitive Data Protection (DLP), IAM, Cloud Functions, Secret Manager
• Vector Search, Vertex AI Search, AlloyDB pgvector, BigQuery vector search
• LangChain, LangGraph
• Gen AI evaluation service, CI/CD
• Terraform, Looker, Looker Studio
• Model Context Protocol (MCP), Agent2Agent (A2A) protocol

Nice to have

• Insurance domain experience (P&C, E&S, specialty, workers’ comp) or financial services.
• Experience with Document AI custom extractors, including OCR for scanned or handwritten forms.
• Experience with Agent2Agent (A2A) protocol or Model Context Protocol (MCP) tool integration.
• Experience with model tuning (supervised fine-tuning on Gemini) and model monitoring.
• Experience with Terraform and dashboards in Looker or Looker Studio.

Preferred certifications

• Google Cloud Professional Machine Learning Engineer
• Google Cloud Associate Cloud Engineer or Professional Cloud Developer
• Google Cloud Generative AI Leader (good to have)

Compensation

USD 105,000 - 115,000 per year (base).

Benefits

• Paid time off based on employee grade (A-F), defined by policy: Vacation 12-25 days depending on grade, company paid holidays, personal days, and sick leave.
• Medical, dental, and vision coverage (or provincial healthcare coordination in Canada).
• Retirement savings plans (for example 401(k) in the U.S. or RRSP in Canada).
• Life and disability insurance.
• Employee assistance programs.
• Other benefits as provided by local policy and eligibility.

Location: Hartford, CT (onsite).