LLM Fine-Tuning - Arabic (Internship)

Company Description

Aralects is developing a gamified, self-paced app focused on teaching Arabic dialects like Egyptian, Lebanese, and Emirati to enhance real-world conversational skills. Personalized learning pathways powered by AI aim to make language learning more engaging and accessible. The company is committed to creating a culturally immersive platform for mastering Arabic dialects, catering to a wide audience for professional growth or personal interest.

Role Description

This is a part-time (or full-time if possible) remote role for a LLM Fine-Tuning Intern at Aralects, and will give the intern the opportunity to be supervised by field experts, add interesting work experience to their CV, and be part of an amazing team that is committed to creating a killer product!

We are seeking a highly motivated intern with experience in both Arabic NLP, Deep Learning, and preferably experience fine-tuning or using LLMs in downstream applications. You will be supervised by the lead scientist and be part of the Aralects team that is composed of product designers, developers, language experts, and scientists.

Key Responsibilities

• Conduct experiments related to fine-tuning LLMs for educational applications.
• Conduct some research to determine best practices and be aware of the state-of-the-art.
• Work with the software development team to integrate that code into the Aralects codebase.

Qualifications

• 2+ years of experience of Python programming experience.
• 1+ years of experience of PyTorch experience (projects, etc.).
• Preferably 1+ years of experience of Arabic NLP experience.
• Optimally has some research experience (familiarity with deep learning and/or Arabic NLP research).
• Is currently working towards a Bachelor's or Master's degree in Computational Linguistics, Machine Learning or AI Engineering, Computer Science, or any related field.

Terms and Compensation

This internship starts as soon as an agreement between the startup and the intern is reached, and it is unpaid for the first two months, pending a satisfying outcome and a smooth relationship, taking into account that we are still an early-stage startup. This is an open internship that is project-based, and if the intern is successful with the first few tasks, then the potential for compensation and even future collaboration will be greatly increased.