Overview

Client: The client is a UK recruitment consultancy. Project Overview: The project covers two use cases: contract and fee extraction from client agreements and client intelligence through meeting notes. DataArt will build a headless agentic backend that handles all reasoning and orchestration. The solution will be built on a custom Python orchestration layer on Cloud Run, calling Vertex AI Search and Gemini on a pay-per-token basis. The backend will expose REST APIs that deliver structured JSONs over BigQuery and document stores, which can be consumed directly by the client’s CRM platform or, optionally, through a custom UI deployed in the client’s GCP environment. The engagement is delivered on a fixed-price basis with milestone-based acceptance.

Responsibilities:
  • Own the solution architecture for the headless agentic backend: custom Python orchestration (ADK/FastAPI) on Cloud Run exposing REST APIs.
  • Design the multi-step 'reason-and-act' logic and agent orchestration (RAG, Text-to-SQL, charting/aggregation) over BigQuery and document stores.
  • Drive direct, pay-per-token integration with Vertex AI Search and Gemini.
  • Lead prompt engineering, routing logic, and design of human-in-the-loop validation workflows.
  • Define evaluation and quality monitoring of AI-generated outputs against the golden dataset; set output quality thresholds for acceptance.
  • Provide technical leadership to the AI/ML engineer; author the architecture, prompt-configuration and scaling-roadmap documentation; support knowledge transfer.
Required Qualifications:
  • Strong Python and hands-on experience building LLM/agentic backends (orchestration frameworks; FastAPI).
  • Deep experience with Vertex AI (Search and Gemini models) and direct LLM API integration; RAG and Text-to-SQL patterns.
  • Prompt engineering, evaluation, and hallucination-mitigation expertise.
  • Experience deploying serverless services on Cloud Run and exposing production REST APIs returning structured JSON.
  • Solid understanding of evaluation methodology and quality monitoring for GenAI outputs.
  • Technical leadership, strong problem-solving and excellent communication skills.
Note:

✨ Our intelligent job search engine discovered this job and republished it for your convenience.
Please be aware that the job information may be incorrect or incomplete. The job announcement remains the property of its original publisher. To view the original job and its full details, please visit the job's URL on the owner’s page.

Please clearly mention that you have heard of this job opportunity on https://ijob.am.