AI Engineer with Gemini and Document AI, Intelligent Automation Platform
About the Position
We are looking for an AI Engineer with strong practical experience in computer vision, image analysis, and Google Cloud AI services. The main focus of the role is to design and implement image- and document-processing workflows using existing Google Cloud APIs and multimodal models. You will analyze photographs, extract visual information, define photo-quality rules, and integrate Google AI services with document-processing and business workflows. The role focuses on selecting, integrating, configuring, and evaluating existing AI services rather than training computer vision models from scratch.
About the Project
This project focuses on building Google Cloud-based intelligent automation solutions for processing business documents and analysing photographic evidence. The solution will support agreed document types, implement document ingestion and structured data extraction, and define rules for assessing photo suitability and quality. The project will primarily use existing Google Cloud services such as Cloud Vision API, Document AI, Gemini on Vertex AI, Cloud Storage, BigQuery, and Cloud Run rather than developing and training custom computer vision models from scratch.
About the Team
You will collaborate with AI engineers, data engineers, cloud engineers, and business analysts in an iterative and collaborative environment. The work will involve close cooperation with business stakeholders to understand document formats, photographic evidence requirements, and business rules for image analysis and quality assessment.
Responsibilities
- Build image-analysis workflows using Google Cloud Vision API and Gemini on Vertex AI.
- Define rules for photo quality, including blur, lighting, resolution, framing, occlusion, and content relevance.
- Process documents and images using Document AI, OCR, and multimodal models.
- Develop Python services and APIs deployed on Cloud Run.
- Integrate Cloud Storage, BigQuery, Pub/Sub, and Eventarc.
- Implement structured AI outputs, confidence scoring, exception handling, and human review.
- Create test datasets and evaluate image- and document-processing accuracy.
Requirements
- Strong knowledge of Python, REST APIs, and FastAPI.
- Practical experience with computer vision and image processing.
- Experience with OpenCV, Pillow, or similar libraries.
- Hands-on experience with Google Cloud Vision API, Vertex AI Gemini, or comparable managed AI services.
- Experience with Document AI, OCR, and structured data extraction.
- Understanding of multimodal LLMs and JSON-schema-based outputs.
- Experience with Cloud Storage, BigQuery, and Cloud Run.
- Ability to select between deterministic image-processing rules, managed vision services, and multimodal AI models.
Nice to Have
- Experience with NumPy, scikit-image, or image-quality metrics.
- Knowledge of Pub/Sub, Eventarc, Docker, Terraform, and CI/CD.
- Experience with human-in-the-loop systems and AI evaluation.
- Understanding of privacy and security requirements for business documents and photographs.
