- Data Analysis and Preprocessing : Analyze and preprocess diverse datasets relevant to the mortgage industry, ensuring data quality and relevance for model training.

- Model Development and Fine-Tuning : Research and implement state-of-the-art NLP models, focusing on pre-training as well instruction tuning pre-trained LLMs for mortgage-specific applications. Utilize techniques like RLHF to improve model alignment with human preferences and enhance decision-making capabilities.

- Algorithm Implementation : Develop and optimize machine learning algorithms to enhance model performance, accuracy, and efficiency. Experiment with different architectures and open-source models to identify the best fit for project requirements.

- Collaboration : Work with domain experts to incorporate industry knowledge into model development, ensuring outputs are relevant and actionable.

- Experimentation : Conduct experiments to validate model hypotheses, analyze results, and iterate on model improvements.

- Documentation : Maintain comprehensive documentation of methodologies, experiments, and results to support transparency and reproducibility.

- Ethics and Bias Mitigation : Ensure responsible AI practices are followed by identifying potential biases in data and models, implementing strategies to mitigate them.

Required Skills :

- Technical Expertise : Strong background in machine learning, deep learning, and NLP. Proficiency in Python and experience with ML frameworks such as TensorFlow or PyTorch.

- NLP Knowledge : Experience with NLP frameworks and libraries (e.g., Hugging Face Transformers) for developing language models.

- Data Handling : Proficiency in handling large datasets, feature engineering, and statistical analysis.

- Problem Solving : Strong analytical skills with the ability to solve complex problems using data-driven approaches.

- Communication : Excellent communication skills to effectively collaborate with technical teams and non-technical stakeholders.