LLM Finetuning
On-siteBengaluru, Karnataka, India
Job Summary
Conduct LLM fine-tuning for financial lending domains including customer onboarding, loan servicing, and collections. Build comprehensive Python modules for NLP tasks such as tokenization, word embeddings, and classifications using TensorFlow, Pytorch, and Transformers. Perform data preprocessing including text cleaning, normalization, and special character handling. Adapt pre-trained language models to specific tasks and domains while optimizing performance via hyperparameter tuning methods like grid search or Bayesian optimization. Evaluate model predictions using appropriate metrics and interpret results. Mitigate bias and preserve privacy in alignment with responsible AI practices. Work with existing customer systems and manage large datasets.
Required Qualifications
- Python Programming: Proficiency in Python
- Working with libraries like TensorFlow, PyTorch, Transformers, NLTK, Pandas, sklearn, and other related libraries used in NLP tasks and fine tuning language models
- Experience with building comprehensive python modules for NLP tasks like tokenization, word embeddings, classifications etc...
- Proficiency in data preprocessing techniques such as text cleaning, tokenization, normalization, handling of special characters is essential for preparing data for fine tuning language models
- Knowledge of machine learning concepts like neural networks, optimization algorithms, model evaluation techniques
- Deep learning expertise w.r.t NLP
- Expertise on adapting pre-trained language models to specific tasks and domains
- Experience with hyperparameter tuning methods such as grid search, random search, or Bayesian optimization for optimizing model performance during fine tuning
- Experience in financial lending domain customer onboarding, loan servicing, collections
- Expertise with tools and frameworks for handling large datasets
- Ability to evaluate model performance using appropriate metrics and interpret model predictions
- Awareness of ethical considerations related to language models, such as bias mitigation, privacy preservation, and responsible AI practices
- Ability to work with existing customer systems
Hiring someone like this?
Get your role in front of qualified candidates on Sorce.