About the job offer

Neuralk-AI is looking for AI scientists that are expert in AI model design and training.

Join our Research & Development team, and contribute to our cutting-edge Applied-AI initiatives.

To build SOTA foundation models for structured data representation, individuals are expected to:

• apply knowledge of relevant research domains or develop novel accurate AI algorithms, along with expert coding skills,
• work towards long-term ambitious research goals, while identifying intermediate milestones,
• collaborate on open-source and scientific breakthroughs by producing publications.

Key Responsibilities:

• Participating to the pre-training effort
• Optimising the architecture of the foundation model
• Designing and consolidating training and evaluation datasets to optimise representational as well as transfer learning abilities of our embedding models
• Running analyses to understand the learning mechanisms of the foundation model
• Working on the decodability of the embedding space

Minimum Qualifications:

• Proven work experience in a university or industry lab(s), in a role with primary emphasis on AI research (transformers/GNN).
• High scientific understanding of the field of AI. This means a broad knowledge of the field of AI, and specific knowledge of interest in training and fine-tuning AI models (transformers / GNNs) with a focus on architecture development, model evaluation and usage.
• High technical engineering competence. This means being able to design complex software and make them usable in production.
• Being able to navigate the full MLOps technical stack.
• Being able to use complex HPC infrastructure with full autonomy.
• Hands-on experience with AI frameworks and tools (e.g., Sklearn, PyTorch, Jax).
• Experience working with large distributed systems.
• Self-starter and autonomous.
• Low-ego.
• Collaborative and have a real team player mindset.

Preferred Qualifications:

• Ph.D. and established track record of leading and/or contributing to influential research, e.g. as evidenced by high-impact publications at peer-reviewed AI conferences (e.g. NeurIPS, CVPR, ICML, ICLR, ICCV, ACL, Interspeech and ICASSP).
• Demonstrated machine learning experience in one of the following: open-source activity, data science competitions.
• Experience in creating high-performance implementations in deep learning frameworks (PyTorch, CUDA) at scale.
• Experience with distributed systems or on-device algorithm development
• 2+ years of work experience in both a faculty and an industry research position with primary emphasis on AI research
• Experience driving original scholarship in collaboration with a team
• Track record of translating research into business impact
• Experience in developing and debugging in C/C++, Python