remote
Freelance Research Engineer - Chart-Grounded Fact Correction - Guires Solutions Private Limited
Research Engineer
Freelance Research Engineer to develop and run end‑to‑end experiments for chart‑grounded fact correction, leveraging Python, PyTorch, computer vision, OCR, and NLP to validate and correct textual claims against visual data.
About the role
Key Responsibilities
- Implement and optimize the chart‑grounded fact correction framework using Python and PyTorch.
- Integrate OCR pipelines to extract textual information from chart images.
- Develop NLP modules to compare extracted data with claim statements and generate corrected claims.
- Run baseline experiments, analyze results, and iterate on model improvements.
- Collaborate with data annotation teams to refine labeling guidelines and ensure high‑quality ground truth.
- Maintain code quality with Git, unit tests, and documentation.
Requirements
- Strong programming skills in Python and experience with deep learning libraries (PyTorch/TensorFlow).
- Hands‑on experience in computer vision and OCR technologies.
- Background in NLP, especially in fact verification or claim analysis.
- Familiarity with version control (Git) and collaborative development workflows.
- Excellent problem‑solving abilities and a proactive, independent work style.
Skills
pythonpytorchcomputer visionnlp