About the role#
This is a full-time research internship for Winter 2027. You will collaborate with our researchers to design and implement novel ideas and help bring models to production. We have openings across several areas, including base model training, retrieval augmented generation, data and evaluation, safety, and finetuning. We are open to research applications in any area relating to large language models.
What you'll do#
- Build and train large language models.
- Focus on projects that expand knowledge in language modeling, evaluation, multimodal models, and optimization.
- Share your results through publications, datasets, and code.
- Contribute to research initiatives that have practical applications in our products.
What you'll need#
- Currently pursuing a PhD in Machine Learning, NLP, Artificial Intelligence, or a related discipline.
- Eligible for work authorization in Canada at the time of hire and able to maintain it throughout the internship.
- Experience with large-scale distributed training, data annotation pipelines, or implementing state-of-the-art ML models.
- Familiarity with autoregressive sequence models like Transformers.
- Programming experience in Python, C, C++, or Lua.
- Familiarity with ML frameworks such as JAX, PyTorch, and TensorFlow.
- Previous experience building systems with machine learning and deep learning techniques.
- Demonstrated research expertise through publications at top-tier venues in machine learning, NLP, AI, computer vision, optimization, computer science, statistics, applied mathematics, or data science.
- Ability to solve analytical problems using quantitative methods.
- Proficiency in handling and analyzing complex, high-dimensional data.
- Experience applying theoretical and empirical research to real-world problems.
- Strong communication skills to explain complex findings clearly.
Location & details#
- Location: Canada
- Term: Winter 2027
- Modality: On-site
- Duration: 4-6 months
- Paid: Yes
About Cohere
Cohere builds foundation models and artificial intelligence products for enterprise use. Founded in 2019, the company focuses on natural language processing and machine learning. It provides tools that allow businesses to deploy AI across private cloud environments or on-premises. The company maintains its headquarters in Toronto and operates several international offices.
How to get in at Cohere
Applying to Cohere early provides a distinct advantage because recruiters review initial submissions before the volume of candidates becomes unmanageable. Intern Insider sends an instant alert the moment a role matching your target is published anywhere, so you apply among the first and ensure your materials are seen before the pile grows. Reaching out to the right people often yields better results than submitting an application into a general queue. Intern Insider surfaces the recruiters behind the company roles so you can reach out directly to ask about the position or a referral, which materially improves response rates.
