Dear all,
We are looking for motivated engineers focusing on reinforcement learning and more specifically on reinforcement learning from human feedback in order to build the RLHF for the Falcon LLM model.
Please find below the complete description of this job opportunity.
Technology Innovation Institute (TII) is a publicly funded research institute, based in Abu Dhabi, United Arab Emirates.
It is home to a diverse community of leading scientists, engineers, mathematicians, and researchers from across the globe, transforming problems and roadblocks into pioneering research and technology prototypes that help move society ahead.
AI and Digital Science Research CentreThe Artificial Intelligence Cross-Center Unit is the machine learning powerhouse of TII, working in close collaboration with our other research centers to harness the full benefits of AI across our projects and drive innovation from new computing paradigms, designing and delivering new AI methodologies, technologies, solutions, and systems that address challenging issues across multiple sectors of the economy from technology to healthcare, cybersecurity, and government, among others.We incorporate core elements of intelligence (perception, sensing, planning, and language) in the ideation design, and prototyping of next-generation systems with human-like intelligence.
We build advanced AI computing and scalable AI-based software stacks and hardware systems to deliver significant enhancements in systems infrastructure. Our AI researchers, scientists, and engineers collaborate to ensure innovative outcomes, from AI theory to AI technologies towards better intelligence.
Job Description
• Familiarity with the paradigm of reinforcement learning from human feedback (RLHF)
• Experience with following and reproducing academic publications
• Experience with any of the following:
• NLP
• Computer Vision
• Large language model training and finetuning
• Development of CUDA drivers for optimized GPU training
• HPC
• Cloud computing (Amazon Web Services, Azure, )• Front-End and/or back-end development• LangChainSkills Required• 3+ Years work experience developing in Python• Strong experience in Deep Learning / ML applications (Pytorch, TensorFlow, )• Strong experience with code versioning and containerization• Strong experience in deep reinforcement learning architecture (DQN, PPO, TRPO, actor critic)
• Experience with distributed training• You look forward to learning with researchers and engineers with diverse backgrounds and expertis• Master's degree and/or PhD in above in a relevant field (e.g., software engineering, computer science)• Knowledge of Arabic is not necessary, but it can be a plus for some projects
Qualifications Required Master's degree and/or PhD in above in a relevant field (e.g., software engineering, computer science)