Machine Learning Intern

Job Description

Machine Learning Intern/Co-op (Fall, 2026) | Cohere

The Tone:
This is an internship at Cohere, a leading security-first enterprise AI company that builds cutting-edge foundation AI models and end-to-end products. Cohere focuses on solving real-world business problems for enterprises by training and deploying frontier models for AI systems. This role is instrumental to the widespread adoption of AI, and interns are expected to contribute to increasing the capabilities of models and the value they drive for customers.

The TL;DR
• Role: Internship
• Type: Full-time
• Location: Remote-friendly, with offices in Toronto, San Francisco, New York City, London, Paris, Montreal
• Mission: Contribute to the design, training, and improvement of cutting-edge AI models to solve real-world business problems.
• Tech Stack: Python, Tensorflow, TF-Serving, JAX, XLA/MLIR, CUDA, TPUs

What You’ll Actually Do
• Design: Design, train, and improve upon cutting-edge models for various applications.
• Develop: Help us develop new techniques to train and serve models in a safer, better, and faster manner.
• Train: Train extremely large-scale models utilizing massive datasets to enhance their capabilities.
• Explore: Explore continual and active learning strategies specifically tailored for streaming data scenarios.
• Collaborate: Work closely with product teams to develop and implement effective machine learning solutions.

The Must-Haves
• Background: Student currently enrolled in a post-secondary program with a demonstrated passion for applied NLP models and products.
• Experience: Experience using large-scale distributed training strategies and familiarity with autoregressive sequence models, such as Transformers.
• Skills: Proficiency in Python and related ML frameworks such as Tensorflow, TF-Serving, JAX, and XLA/MLIR, combined with strong communication and problem-solving abilities.
• Bonus: Experience writing kernels for GPUs using CUDA, experience training on TPUs, or papers published at top-tier venues (such as NeurIPS, ICML, ICLR, AIStats, MLSys, JMLR, AAAI, Nature, COLING, ACL, EMNLP).