AI Engineer
Role Overview
This is an offshore AI Engineering role embedded within the Azuria estimation team to enhance and extend an existing AI-powered map extraction system. The engineer will work closely with estimation leads and cross-functional stakeholders to improve model accuracy, expand extraction capabilities, and optimize end-to-end SaaS workflows. The ideal candidate brings strong applied ML and computer vision experience, comfort working within an existing codebase, and the ability to operate with minimal supervision in a collaborative, async-first environment.
Key Responsibilities
- Customize and fine-tune large language models (LLMs) and computer vision models to improve map feature extraction accuracy across diverse input formats (PDFs, scanned drawings, GIS exports)
- Extend and optimize the existing AI map extraction pipeline — including pre-processing, inference, post-processing, and output validation stages
- Collaborate with Azuria's estimation team to translate domain requirements into model improvements and feature enhancements
- Design and implement evaluation frameworks (precision, recall, IoU, etc.) to measure model performance and guide iteration
- Build and maintain data engineering pipelines for training data ingestion, labeling quality assurance, and dataset versioning
- Optimize SaaS platform workflows — reducing latency, improving throughput, and enhancing reliability of AI-driven outputs within the production application
- Participate in code reviews, technical documentation, and knowledge-sharing with the broader engineering and estimation teams
- Monitor model performance in production, identify drift or degradation, and implement retraining or remediation workflows
Required Qualifications
- 4+ years of hands-on experience in applied machine learning or AI engineering, with demonstrated work in computer vision or document understanding
- Strong proficiency in Python and ML frameworks (PyTorch, TensorFlow, or equivalent)
- Experience working with LLMs (fine-tuning, prompt engineering, RAG architectures)
- Familiarity with image processing libraries (OpenCV, Pillow) and geospatial or CAD data formats is a strong plus
- Experience with cloud platforms (AWS, GCP, or Azure) and containerized deployment (Docker, Kubernetes)
- Solid understanding of data engineering fundamentals — ETL pipelines, data validation, and storage best practices
- Ability to work independently in an async-first, distributed team environment with minimal supervision
- Strong written communication skills for technical documentation and cross-functional collaboration
Preferred Qualifications
- Prior experience in construction, infrastructure, or utility estimation domains
- Familiarity with GIS tools, spatial data, or map digitization workflows
- Experience building or maintaining evaluation and monitoring frameworks for ML models in production
- Track record of contributing to SaaS products with AI/ML components