Member of Technical Staff, Agent Code

Requirements

• A PhD in Computer Science, Machine Learning, or a related field, with publications in top-tier venues (e.g., NeurIPS, ICML, ICLR, ACL, EMNLP)
• Deep expertise in code LLMs and agent systems, with a strong understanding of the latest research and trends. We are looking for people who not only have worked with code models, but have actively contributed to their development
• Hands‑on experience with frontier LLMs and their applications in code generation or automation
• Strong software engineering skills, with proficiency in Python and PyTorch, TensorFlow, or similar frameworks
• Experience with distributed systems, cloud infrastructure, and scalable architectures
• A proactive, self‑motivated mindset, with a passion for solving ambitious, open‑ended problems
• If you're passionate about shaping the future of code LLMs and agent systems, and thrive in a dynamic, research‑driven environment, we'd love to hear from you!
• If some of the above doesn't line up perfectly with your experience, we still encourage you to apply!

What the job involves

• Code‑generating LLMs and autonomous agents are revolutionizing how software is built and tasks are automated
• At Cohere, we're pushing the boundaries of what's possible with these technologies for enterprises, and we're looking for a senior member for the Agent Code team
• You'll be at the forefront of research and engineering, driving the development of cutting‑edge code LLMs and agent systems that can interact with the digital world to solve complex tasks with minimal human oversight
• This role is hands‑on and research‑driven
• You'll dive into the latest literature on code LLMs and agents, experiment with frontier models, and collaborate with a team of talented engineers and researchers to build scalable, production‑ready solutions
• At Cohere, we blend engineering and research seamlessly—everyone contributes to both, depending on their interests and organizational needs
• We provide access to world‑class compute resources, data, and talent to ensure you can do your best work
• Stay up‑to‑date with the latest research in code LLMs, agents, and related fields, implementing novel ideas into our systems
• Design and implement scalable strategies to train code models, and deploy agent frameworks for inference and sampling. You will be collaborating with the pretraining team, creating SFT trajectories and working on existing and new RL algorithms
• Hillclimb on existing benchmarks and design new ones that reflect the needs of our enterprise users
• Lead experiments on our state‑of‑the‑art compute infrastructure, pushing the boundaries of what's possible with frontier LLMs