Audio Inference Engineer, Model Efficiency

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 co-headquartered in Toronto and San Francisco, with key offices in London, New York City, Montreal, Seoul, Germany and Paris. Join us! Why this role? Our team is a fast-growing group of committed researchers and engineers. The mission of the team is to build reliable machine learning systems and optimize audio inference serving efficiency using innovative techniques. As an engineer on this team, you will work on advancing core audio model serving metrics, including latency, throughput, and quality by diving deep into our systems, identifying bottlenecks, and delivering creative solutions for audio processing and streaming workloads. You’ll collaborate closely with both the training and serving infrastructure teams to ensure seamless integration between model development and deployment, with a special focus on real-time and streaming audio inference. Please Note: We have offices in Toronto, Montreal, San Francisco, New York, Paris, Seoul and London. We embrace a remote-friendly environment, and as part of this approach, we strategically distribute teams based on interests, expertise, and time zones to promote collaboration and flexibility. You'll find the Model Efficiency team concentrated in the EST and PST time zones, these are our preferred locations. You may be a good fit for the team if you have: - Significant experience developing high-performance audio or machine learning inference systems. - Proficiency with programming languages such as C++ and Python. - Hands-on experience with deep learning models for audio, speech, or language applications. - A bias for action and a strong results-oriented mindset. It is a big plus if you also have considerable experience with: - GPU programming, low-level system optimization, model parallelization techniques over multiple GPUs - Have experience with duplex real-time streaming architectures. - Internals of machine learning frameworks for audio (such as PyTorch, TensorFlow, or specialized audio libraries). - Have experience with inference framework like vLLM, SGLang, Tensort-LLM, or custom distributed inference systems - Sequence modeling (e.g., transformers for audio/speech) and end-to-end audio pipeline optimization Full-Time Employees at Cohere enjoy these Perks: - A weekly lunch stipend of $75/£75 or equivalent in your local currency for lunch. - Full health and dental benefits, including a separate budget for mental health. - RRSP matching, 401K, Pension Scheme. - 100% Parental Leave top-up for up to 6 months, for either parent. - Annual enrichment benefits: Arts & culture, fitness/wellness, quality time, and a workspace improvement credit. Education & learning stipend for conferences, courses, and coaching. - 6 weeks of paid vacation (30 working days!) - Budget for traveling to other offices if you are remote, plus an annual company offsite. How and Where We Work: - Cohere is remote-friendly. We have offices in Toronto, San Francisco, New York City, London, Paris, Montreal, and more coming soon. - For those in the office: a daily lunch program, plenty of snacks, and regular community and social events. - For those not near an office: a co-working benefit so you can work alongside others in your city. - Everyone receives a $500 home office stipend to set up your workspace properly. If any of the above doesn’t line up exactly with your experience, we still encourage you to apply. We strive to create an inclusive work environment for all; we welcome applicants from all backgrounds and are committed to providing equal opportunities. Should you require any accommodations during the recruitment process, please submit an Accommodations Request Form, and we will work together to meet your needs. We may use AI-enabled tools to screen and assess applicants against the criteria for this position. This helps our recruiters identify potentially qualified candidates, but it doesn't limit the applications our recruiters may review or consider. Beware of Scams: Cohere will never ask for payment or third-party services (e.g., CV writing) as part of our hiring process. All legitimate roles are listed on the Cohere careers page and LinkedIn only, with all communications from Cohere employees coming from an @cohere.com or @cw.cohere email a

Significant experience developing high-performance audio or machine learning inference systems. Proficiency with C++ and Python. Hands-on experience with deep learning models for audio, speech, or language applications. A bias for action and a strong results-oriented mindset. Bonus: GPU programming, low-level system optimization, model parallelization over multiple GPUs; experience with duplex real-time streaming architectures; familiarity with ML frameworks for audio (e.g., PyTorch, TensorFlow) or specialized audio libraries; experience with inference frameworks like vLLM, SGLang, Tensort-LLM, or custom distributed inference systems; experience with sequence modeling (e.g., transformers for audio/speech) and end-to-end audio pipeline optimization.

Build and optimize reliable machine learning systems for audio inference and streaming workloads. Improve core audio model serving metrics (latency, throughput, quality). Identify bottlenecks in the system and deliver solutions for real-time and streaming audio inference. Collaborate with training and serving infrastructure teams to ensure seamless integration between model development and deployment.