Key Responsibilities: Include collaborating closely with research scientists and other team members to explore and use advanced machine learning techniques in NMT within the scope of the project, like translation of Indian and foreign languages across various domains such as legal, entertainment, academic, etc. Interest or prior exposure to at least one of the following is desirable:
• Domain adaptation & Out-of-Distribution robustness
• Efficient adaptation of LLMs
• Mixture of Experts (MoE) architectures
• Representation Learning
Work Location: Remote within India
Duration of the paid internship: The paid internship will be for 6 months starting from first week of October 2025.
• The working hours are from 9:00 to 18:00 (Monday to Friday) full-time.
Essential Education: Ph.D. or Master’s with a Computer Science background or Bachelor’s with substantial hands-on Python experience applied to LLMs/NLP.
Skills required:
Essential Skills: Excellent coding skills, especially in Python and PyTorch
• Knowledge of the state-of-the-art in LLMs/Foundation Models/NMT
• Basics of Mixture-of-Experts (MoE)
• Natural Language Processing
Good to have Skills: Strong mathematical aptitude
• Experience in domain adaptation techniques.
• Papers in top-tier conferences like ICML, NeurIPS, ICLR, AAAI, ACL, etc.