Back to Jobs
Recently postedTechnology

AI/ML Data architect

Saim TechnologiesUnited States🇺🇸United States

Quick overview

SeniorityMid Senior
Work modeHybrid
LocationUnited States
DockerMongoDBSQLShellGitApache SparkETLMLOpsMLflowNLPNumPyOpenCVScikit-learnBashComputer VisionConfluenceDatabricksDeep LearningGitHub ActionsHugging FaceLLMPandasPyTorchPythonTensorFlowGenerative AIRESTAzurePostgreSQLKubernetesJavaScriptTypeScriptMachine LearningSQL ServerJiraPrompt EngineeringSharePointCI/CDAzure DevOpsAzure Data FactoryDelta LakeLangChainLangGraphRAG

Job description

JD

Artificial Intelligence & Machine Learning

Programming & Scripting

Azure OpenAI

Hugging Face Models

AI Agent Frameworks

Retrieval-Augmented Generation (RAG)

Document Intelligence & OCR

Natural Language Processing (NLP

Job Summary

Project 1: Module Prints – AI-Powered Engineering Diagram Analysis

The Module Prints project focuses on using Artificial Intelligence to automatically extract relevant technical information from engineering diagrams and module prints. The solution applies computer vision and machine learning techniques to identify components, annotations, dimensions, labels, and other engineering-specific information contained within complex drawings. The goal is to convert traditionally unstructured visual engineering data into structured, searchable information that can be used by downstream applications and engineering teams. Key skills involved in this project include computer vision, image processing, machine learning algorithms, object detection, and AI-based information extraction.

Project 2: IDI – Intelligent Document Intelligence

The Intelligent Document Intelligence (IDI) project uses AI to extract and interpret information from complex PDF documents related to vehicle accidents. The system is designed to identify important details such as the vehicles involved, accident location, crash site information, potential causes of the accident, contributing factors, and other relevant incident data. AI-based document understanding techniques are used to process both textual and visual content within PDFs and transform the extracted information into structured data for further analysis. Key technologies include document AI, natural language processing, computer vision, machine learning, OCR, and intelligent information extraction.

Project 3: AI Agent for Codebeamer Requirement Enrichment

This project involves developing an intelligent AI agent that assists engineers in creating and improving requirements within Codebeamer. The agent analyzes existing requirement statements, associated documentation, engineering diagrams, and relevant organizational knowledge to identify missing information, improve clarity, and enrich requirements with additional technical context. Advanced Retrieval-Augmented Generation (RAG) techniques can be used to retrieve relevant historical requirements, standards, specifications, and supporting documentation before generating recommendations. Computer vision capabilities can also enable the agent to understand diagrams and other visual engineering artifacts associated with a requirement. Key skills include advanced RAG architectures, AI agents, large language models, computer vision, semantic search, vector databases, prompt engineering, and enterprise knowledge integration.

Key Responsibilities

Project 1: Module Prints – AI-Powered Engineering Diagram Analysis

The Module Prints project focuses on using Artificial Intelligence to automatically extract relevant technical information from engineering diagrams and module prints. The solution applies computer vision and machine learning techniques to identify components, annotations, dimensions, labels, and other engineering-specific information contained within complex drawings. The goal is to convert traditionally unstructured visual engineering data into structured, searchable information that can be used by downstream applications and engineering teams. Key skills involved in this project include computer vision, image processing, machine learning algorithms, object detection, and AI-based information extraction.

Project 2: IDI – Intelligent Document Intelligence

The Intelligent Document Intelligence (IDI) project uses AI to extract and interpret information from complex PDF documents related to vehicle accidents. The system is designed to identify important details such as the vehicles involved, accident location, crash site information, potential causes of the accident, contributing factors, and other relevant incident data. AI-based document understanding techniques are used to process both textual and visual content within PDFs and transform the extracted information into structured data for further analysis. Key technologies include document AI, natural language processing, computer vision, machine learning, OCR, and intelligent information extraction.

Project 3: AI Agent for Codebeamer Requirement Enrichment

This project involves developing an intelligent AI agent that assists engineers in creating and improving requirements within Codebeamer. The agent analyzes existing requirement statements, associated documentation, engineering diagrams, and relevant organizational knowledge to identify missing information, improve clarity, and enrich requirements with additional technical context. Advanced Retrieval-Augmented Generation (RAG) techniques can be used to retrieve relevant historical requirements, standards, specifications, and supporting documentation before generating recommendations. Computer vision capabilities can also enable the agent to understand diagrams and other visual engineering artifacts associated with a requirement. Key skills include advanced RAG architectures, AI agents, large language models, computer vision, semantic search, vector databases, prompt engineering, and enterprise knowledge integration.

Skill Requirements

Consolidated Technology Stack

Programming & Scripting

· Python

  • SQL
  • TypeScript / JavaScript
  • REST APIs
  • Bash / Shell Scripting

Artificial Intelligence & Machine Learning

· Machine Learning

  • Deep Learning
  • Generative AI
  • Large Language Models (LLMs)
  • Multimodal AI
  • Prompt Engineering
  • Fine-Tuning & Model Evaluation

LLM Platforms

· Azure OpenAI

  • Anthropic Claude
  • Llama 3.x
  • Hugging Face Models

AI Agent Frameworks

· Semantic Kernel

  • LangChain
  • LangGraph
  • AutoGen
  • CrewAI
  • Agentic AI Design Patterns
  • Function Calling & Tool Integration

Retrieval-Augmented Generation (RAG)

· Advanced RAG Architectures

  • Graph RAG
  • Hybrid Search
  • Semantic Search
  • Knowledge Graphs
  • Context Enrichment
  • Enterprise Knowledge Retrieval

Document Intelligence & OCR

· Azure AI Document Intelligence

  • OCR Technologies (Azure OCR, Tesseract, EasyOCR, PaddleOCR)
  • PDF Processing (PyMuPDF, PDFMiner, pdfplumber)
  • Document Classification
  • Information Extraction
  • Form Recognition
  • Layout Analysis

Natural Language Processing (NLP)

· spaCy

  • Hugging Face Transformers
  • NLTK
  • Named Entity Recognition (NER)
  • Text Classification
  • Entity Extraction
  • Summarization
  • Semantic Similarity

Computer Vision & Image Processing

· OpenCV

  • PIL/Pillow
  • Azure AI Vision
  • Object Detection
  • Image Segmentation
  • Engineering Diagram Analysis
  • Technical Drawing Interpretation
  • Annotation & Symbol Detection
  • Visual Data Extraction

Deep Learning Frameworks

· PyTorch

  • TensorFlow
  • Scikit-learn
  • Detectron2
  • YOLO
  • Vision Transformers (ViT)
  • LayoutLM
  • Donut
  • DocFormer

Vector Databases & Search

· Azure AI Search

  • Pinecone
  • Weaviate
  • FAISS
  • ChromaDB
  • Milvus
  • Elasticsearch
  • OpenSearch

Data Engineering & Analytics

· Pandas

  • NumPy
  • PySpark
  • Databricks
  • ETL Pipelines
  • Data Modeling
  • Data Lakes
  • Delta Lake

Databases

· SQL Server

  • PostgreSQL
  • Cosmos DB
  • MongoDB

Enterprise Integrations

· Codebeamer

  • Azure DevOps
  • GitHub
  • Jira
  • Confluence
  • SharePoint
  • Engineering Knowledge Repositories

Cloud & MLOps

· Microsoft Azure

  • Azure Machine Learning
  • Azure Functions
  • Azure Storage
  • Azure Kubernetes Service (AKS)
  • Azure Data Factory
  • MLflow
  • Docker
  • Kubernetes
  • CI/CD Pipelines
  • GitHub Actions
  • Azure DevOps Pipelines

Domain Expertise

· Automative Industry

  • Requirement Engineering
  • Requirement Traceability
  • Requirement Enrichment
  • Enterprise Knowledge Management

AI-Powered Information Extraction

Similar jobs