AI Research Engineer, Model Compression – Quantization

This is a fully remote position, open to applicants in Switzerland.

📋 Description

• Lead the way in innovating model compression and efficient deployment for cutting-edge multimodal AI systems, including large language models (LLMs) and vision-language models (VLMs).

• Aim to minimize model size and computational expenses while maintaining accuracy.

• Utilize and enhance compression techniques such as quantization, knowledge distillation, and pruning.

• Create, evaluate, and implement novel strategies for compression that balance model size, latency, throughput, and accuracy.

• Establish robust compression pipelines, defining performance and fidelity metrics, and resolving bottlenecks in production inference.

⛳️ Requirements

• A degree in Computer Science or a related discipline.

• Preferably a PhD in NLP, Machine Learning, or a related area, supported by a strong record in AI research and development (with notable publications in A* conferences).

• Proficiency with PyTorch deep learning frameworks or equivalent alternatives.

• Practical experience with model quantization, including both Quantization-Aware Training (QAT) and Post-Training Quantization (PTQ).

• Research and practical experience in knowledge distillation for transforming large models into smaller, more efficient versions.

• Research and practical experience in model pruning for compressing large models into smaller, efficient iterations.

• A solid grasp of neural network architectures and training methodologies – including transformers (e.g., LLMs, VLMs), backpropagation, optimization, and fine-tuning techniques.

• Familiarity with C++ is advantageous (especially for implementing low-level quantization kernels or enhancing inference optimizations).

🏝️ Benefits

• Our team represents a global talent pool, collaborating remotely from various locations worldwide.

• Engage with some of the brightest minds in the industry, challenging limits and establishing new benchmarks.

• Chance to contribute to the most innovative platform on the planet.