Senior GenAI Data Integration Engineer

We are seeking a highly skilled and dedicated Senior GenAI Data Integration Engineer to lead the integration of a key client’s advanced Generative AI capabilities (Gemini on Vertex AI) with our core Business Intelligence platform, Looker. This is a critical, dedicated role focused on transforming how our business users interact with data, moving from static reports to dynamic, conversational, and explanatory data narratives. The ideal candidate must be an expert in both LookML and LLM integration/prompt engineering, capable of building and deploying production-ready, well-documented, and reusable conversational analytics frameworks using exclusively Google tools and compliant languages.

This is a 6-month dedicated contract role, with the strong possibility of a longer-term extension based on project success and future strategic needs.

The successful candidate will be fully responsible for the end-to-end development, deployment, and documentation of three core capabilities, utilizing only Google Cloud Platform (GCP) services and standard compliant languages (Python/JavaScript):

●        Design and implement a robust integration layer between Looker's data models and the Gemini API (via Vertex AI).

●        Develop and refine advanced Prompt Engineering strategies to analyze Looker dashboard data, trends, and business drivers, ensuring contextual and accurate narrative output.

●        Automate the generation of high-quality, natural language summaries and narratives explaining specific dashboard trends (e.g. "Narrate the reason for the decrease in Total Approved Cost this month").

●        Engineer and deploy a conversational chatbot interface using JavaScript/React and hosted on GCP (e.g. Cloud Run/App Engine).

●        Utilize Python and the Looker API to effectively guide the LLM in accurately determining the user's intent, translating the request into the necessary Looker filter and view adjustments.

●        Orchestrate a seamless workflow where a user's conversational request triggers a dashboard change, followed by a data-driven narrative.

●        Build a scalable Natural Language Query (NLQ) layer that uses Gemini's function calling or grounding features to translate complex user questions into accurate, governed LookML or BigQuery SQL queries.

●        Implement fine-tuned prompt templates that ground the LLM's output using Looker's semantic layer to ensure query security and data integrity within the BigQuery environment.

●        Establish transparent, modular, and version-controlled code repositories (e.g. Cloud Source Repositories or standard Git) for all integration logic (Python/JavaScript).

●        Create comprehensive, well-structured documentation for all components: Prompt Engineering library, Looker API interface, and deployment configuration on GCP services.

●        Ensure the solution is built with modularity using Google-compliant patterns to allow easy expansion and maintenance by a future team.

●        Generative AI / LLM Expertise: Proven experience developing applications using Gemini Pro/Flash or other models within the Vertex AI platform. Expertise in Prompt Engineering for BI tasks.

●        Business Intelligence Expertise: Expert-level proficiency in Looker or similar tools, including advanced LookML data modeling and Looker API integration.

●        Cloud & Data Architecture: Strong working knowledge of Google Cloud Platform (GCP), specifically BigQuery, Vertex AI, Cloud Run/App Engine, and Cloud Functions.

●        Programming & APIs: Expert proficiency in Python for API orchestration and backend logic. Strong skills in JavaScript for frontend integration. Experience with version control (Git).

●        Data Governance: Strong understanding of data security principles, including Looker's Row-Level Security and Access Filters over BigQuery data.

●        Successful, production-ready deployment of the Conversational BI Chatbot for key operational dashboards, running entirely on GCP infrastructure.

●        A complete, fully documented, and handover-ready technical framework built using Python and JavaScript, with all code committed to a version control system.

●        A fully documented and maintainable Gemini/Looker integration framework, including a library of high-performing prompt templates.

●        Measurable increase in data engagement and self-service analytics adoption across target business units.