Research Internship (Winter 2027)

Who are we? Cohere is the leading security-first enterprise AI company. We build cutting-edge foundation AI models and end-to-end products that are designed to solve real-world business problems. We’re training and deploying frontier models for enterprises who are building AI systems. We believe that our work is instrumental to the widespread adoption of AI and we are looking for folks that want to be part of that. We obsess over what we build. Each one of us is responsible for contributing to increasing the capabilities of our models and the value they drive for our customers. Cohere is a team of researchers, engineers, designers, and more, who are all passionate about their craft. We are a global technology company headquartered in Toronto with key offices in London, New York City, San Francisco, Montreal, Paris, Berlin and Seoul. Join us! Why this role? To have the opportunity to collaborate with Cohere researchers and tools on designing and implementing novel research ideas and shipping state-of-the-art models to production. We have openings in teams covering base model training, retrieval augmented generation, data and evaluation, safety, and finetuning, to name a few; and we are open to receiving intern applications in any research area relating to LLMs to broaden your research connections while obtaining deep experience in a growing AI startup. Please Note: To be eligible for a Research Internship, you must be currently pursuing a PhD in Machine Learning, NLP, or a related discipline. You need to be available for a full-time internship that lasts for 4-6 months. Areas of Research We're interested in connecting with researchers working in areas including, but not limited to: Model Performance: Improving model capabilities, quality, reliability, and performance across different tasks and benchmarks. Model Efficiency: Optimizing model architectures, training, and inference for greater computational and memory efficiency. Model Evaluation: Developing evaluation methodologies, benchmarks, and frameworks to measure model capabilities, robustness, and reliability. Post-Training & Reinforcement Learning: Advancing model capabilities through supervised fine-tuning, preference optimization, reinforcement learning, and related techniques. Code Generation & Agentic Coding: Developing models capable of code generation, software engineering tasks, and autonomous coding workflows. Multimodal Learning: Researching models that understand, generate, and reason across multiple modalities, including text, images, audio, and video. Pre-Training, Data & Synthetic Data: Improving pre-training methodologies, data quality, data selection, and synthetic data generation. Mid-Training: Exploring intermediate training strategies to enhance model capabilities and support specialization. AI Safety: Advancing model safety, alignment, robustness, and responsible AI development. We're also open to exploring other research directions related to LLMs and generative AI. As a Cohere Research Intern, you will: Conduct cutting-edge machine learning research, building and training large language models. Focus on research projects aimed at expanding the frontier of knowledge in language modeling and related areas such as evaluation, multimodal models, and optimization. Disseminate your research results through publications, datasets, and code. Contribute to research initiatives with practical applications in Cohere's product development. You may be a good fit if you: Are currently pursuing, or in the process of obtaining, a PhD in Machine Learning, NLP, Artificial Intelligence, or a related discipline. Are eligible for work authorization in the country of employment at the time of hire and can maintain work authorization throughout the internship. Have experience with large-scale distributed training strategies, data annotation and evaluation pipelines, or implementing state-of-the-art ML models. Are familiar with autoregressive sequence models, such as Transformers. Have strong communication and problem-solving skills, with the ability to convey complex research findings clearly and succinctly. Have programming experience in Python, C, C++, Lua, or related languages. Are familiar with ML frameworks such as JAX, PyTorch, and TensorFlow. Have previous experience building systems using machine learning and deep learning techniques. Demonstrate a passion for applied NLP models and products. Preferred Qualifications Demonstrated research expertise through publications at top-tier venues in fields such as machine learning, NLP, artificial intelligence, computer vision, optimization, computer science, statistics, applied mathematics, or data science. Proven ability to tackle analytical problems using quantitative methodologies. Proficiency in handling and analyzing complex, high-dimensional data from various sources. Experience applying theoretical and empirical research to real-world problems.