Responsibilities
- Assist in developing a secure, AI-powered platform for verification and assurance.
- Contribute to developing and improving computer vision models for image forgery detection, replay detection, and advanced fraud analysis.
- Implement and experiment with image processing techniques such as noise analysis, Error Level Analysis (ELA), blur detection, and frequency-domain features.
- Support OCR pipelines using tools like PaddleOCR or Azure AI Vision to extract text from photos of identity documents.
- To help prepare and clean real-world image datasets, including handling low-quality, noisy, and partially occluded images.
- Integrate trained models into Python-based APIs (FastAPI) for internal testing and production use.
- Collaborate with senior engineers to test, debug, and optimise model performance.
- Document experiments, findings, and implementation details clearly.
Requirements
- Proficiency in Machine Learning, Computer Vision, and Python.
- Core experience with OpenCV, NumPy and core image processing concepts.
- Introductory experience with PyTorch or TensorFlow.
- Understanding of Convolutional Neural Networks (CNNs) fundamentals.
- Hands-on experience with REST APIs or FastAPI.
- Exposure to OCR, document processing, or facial analysis.
- 2-4 years of experience in computer vision, image processing, and AI.
- Bachelor's degree in Computer Science, IT, Electronics, or a related discipline.
- Prior experience related to image forensics, fraud detection, and biometrics.
- Comfortable working with imperfect real-world data.
- Good communication skills and a team-oriented mindset.
This job was posted by Rajesh Kumar Goel from Infinity Assurance Solutions.
