Multimodal AI Engineer for Engineering Drawings & Data
Posted 7
views
Summary
This engineer builds multimodal AI systems that understand, search, and extract knowledge from engineering drawings and technical documents, using computer vision, OCR, vision-language models, embeddings, and LLMs. Solutions are built in Python and deployed on AWS (Bedrock, Terraform) and Databricks, with agentic/MCP-based integrations into enterprise engineering systems.
- Develop multimodal AI solutions for engineering drawings, specifications, standards, and related documents.
- Build drawing understanding capabilities using computer vision, OCR, vision language models, embeddings, and LLMs.
- Implement drawing search, retrieval, and knowledge extraction capabilities.
- Design agentic workflows and MCP based integrations across engineering systems.
- Integrate AI solutions with Databricks, cloud platforms (AWS), and enterprise data sources.
- Evaluate emerging AI technologies and recommend architecture decisions.
Required Skills
- AWS services, including AgentCore, Bedrock, and Terraform for application deployment and integration.
- Databricks Apps, Unity Catalog for governed data access, and platform integration.
- Computer vision and multimodal AI for drawing understanding, symbol detection, OCR, and vision language model-based interpretation.
- Python based AI application development, including backend services, APIs, evaluation pipelines, and integration with model inference workflows.
- RAG, vector databases, embeddings, similarity search, and drawing retrieval across engineering drawings, specifications, and related technical documents.
- Multiagent workflow and MCP architecture
- API and enterprise system integration
- Engineering or manufacturing domain experience