Machine Learning Engineer (Document Intelligence & OCR)

Experience : 47 Years | Location: Noida (Hybrid)

Impact at VELLOE

As a Machine Learning Engineer (Document Intelligence & OCR), you will build AI-powered OCR and intelligent document processing solutions using Computer Vision, Deep Learning, and Document Intelligence.

You will have hands-on experience in designing, training, and deploying CNN and Vision Transformer (ViT) models for document classification, information extraction, fraud detection, and OCR of financial and identity documents.

Job Responsibilities :

In this role, you will :

- Design, train, and deploy Machine Learning and Deep Learning models.

- Develop Document Intelligence and OCR solutions for paystubs, bank statements, invoices, tax forms, and identity documents.

- Build and optimize CNN and Vision Transformer (ViT) models for document classification, layout analysis, object detection, and text extraction.

- Develop scalable AI services using Python, FastAPI, or Flask.

- Improve model accuracy, inference performance, and production scalability.

- Collaborate with engineering teams to deploy AI solutions in cloud environments.

Required Qualifications :

The top candidate will have :

- Programming: Python, FastAPI, Flask.

- Machine Learning: Scikit-learn, XGBoost.

- Deep Learning: PyTorch, TensorFlow, Keras.

- Computer Vision: CNN, Vision Transformers (ViT), OpenCV, YOLO.

- Document Intelligence: OCR, Intelligent Document Processing (IDP), Layout Analysis, Key-Value Extraction.

- Cloud & DevOps: Docker, Git, AWS/Azure.

- Databases: PostgreSQL, MongoDB, Redis.

Bonus Qualifications

Ideally you have :

- Generative AI: Hugging Face, LLMs, LangChain, RAG.

Ideal Candidate :

- Hands-on experience designing, training, and deploying CNN and Vision Transformer (ViT) models.

- Comfortable working across document classification, information extraction, fraud detection, and OCR of financial and identity documents.

Key Skills :

Machine Learning, Deep Learning, Python, Computer Vision, CNN, Vision Transformer (ViT), OCR, Document Intelligence, Intelligent Document Processing (IDP), PyTorch, TensorFlow, OpenCV, YOLO, FastAPI, Flask, Scikit-learn, Hugging Face, LLM, LangChain, RAG, Docker, PostgreSQL, MongoDB, Redis.