Staff Research Engineer, Model Efficiency
Toronto, ON +1Cohere: Cohere is the leading security-first enterprise AI company.
Candidates need a PhD in machine learning or a related field, an understanding of LLM architecture and inference optimization, and significant experience with model efficiency techniques. Strong software engineering skills, top-tier publications (ICLR, ACL, NeurIPS), and a willingness to mentor are also required.
Research, Collaboration, Machine learning
Member of Technical Staff, Agent Code
Toronto, ONCohere: Cohere is the leading security-first enterprise AI company.
Cohere needs candidates with a PhD in computer science, machine learning or a related field, with top-venue publications, deep code LLM and agent expertise, and strong Python and PyTorch or TensorFlow skills. Experience with distributed systems and cloud infrastructure is also required.
PyTorch, TensorFlow, Python
Member of Technical Staff, Integration/RL Team (Research Engineer)
Toronto, ON +1Cohere: Cohere is the leading security-first enterprise AI company.
Candidates need very strong software engineering skills, proficiency in Python and ML frameworks such as JAX, PyTorch or XLA/MLIR, and experience debugging large-scale distributed training. Test-driven development and clean code habits are valued. Experience with Kubernetes, Ray, post-training and ML or RL research are bonus qualifications.
Python, JAX, PyTorch, XLA/MLIR, Kubernetes, Ray
Member of Technical Staff, Pre-Training Data
Toronto, ON +1Cohere: Cohere is the leading security-first enterprise AI company.
Candidates need strong Python and data pipeline engineering skills, familiarity with data mixing and data processing frameworks, and experience with large-scale web, code and multilingual datasets. A research-oriented mindset bridging research and engineering is valued.
Python, Apache Spark, Apache Beam, Pandas
Member of Technical Staff, Data Analysis and Evaluation
Toronto, ON +1Cohere: Cohere is the leading security-first enterprise AI company.
Cohere needs candidates with extremely strong software engineering skills, statistical expertise in evaluating experiments, and experience designing data collection with human annotators. Candidates also need hands-on LLM training on distributed infrastructure, proficiency in Python and ML frameworks, and strong communication skills. One or more papers at top-tier venues are listed as a qualification.
Python, PyTorch, TensorFlow, JAX
Member of Technical Staff, Training Performance Engineer
Toronto, ON +1Cohere: Cohere is the leading security-first enterprise AI company.
Candidates need extremely strong software engineering skills, proficiency in Python and ML frameworks such as JAX, PyTorch and XLA/MLIR, and experience writing GPU kernels with CUDA or Triton. Experience with large-scale distributed training and transformer sequence models is also required.
CUDA, Triton, PyTorch, JAX, XLA, MLIR
Senior Member of Technical Staff, Multimodal AI
Toronto, ON +1Cohere: Cohere is the leading security-first enterprise AI company.
Candidates need exceptional software engineering skills, strong Python, and experience with deep learning frameworks such as JAX, PyTorch and TensorFlow. Experience with distributed training, autoregressive models and evaluating large multimodal models is also expected.
Python, JAX, PyTorch, TensorFlow, CUDA, HuggingFace