Senior Technical Program Manager, Machine Learning Infrastructure

About the Role

Cohere is seeking an exceptional Senior Technical Program Manager (TPM) to manage the growing size and complexity of its programs, specifically focusing on Machine Learning Infrastructure. TPMs at Cohere are operational experts specializing in cross-functional efforts, executing with precision tactically while partnering with senior leaders strategically. This role involves managing the program portfolio for inference, efficiency, serving, and endpoints, ensuring Cohere’s infrastructure scales for a rapidly expanding user base. The TPM will lead end-to-end coordination and execution, collaborate with Modeling and customer-facing teams, identify pain points, establish processes for development focus, build a culture of continuous improvement, and manage overlapping projects. This is a dynamic role for someone excited about solving novel problems and bringing structure to a fast-paced environment.

Responsibilities

• Manage the program portfolio covering inference, efficiency, serving, and endpoints to guarantee that Cohere’s infrastructure continues to scale for a rapidly expanding user base of internal and external users
• Lead the end-to-end coordination and execution across the program, with a particular focus on cross-functional collaboration with Modeling and customer-facing teams
• Identify pain points and establish processes to ensure that the team can focus on development, while meeting needs of internal and external users of the models, and improving engineering best practices
• Build a strong culture around continuous improvement, such as liaising with incident management leads, ensuring that problems are accurately root caused, and that required fixes are provided in a timely manner
• Manage various overlapping projects and programs, ruthlessly prioritizing asks, to ensure that the company’s top priorities are met
• Collaborate with stakeholders across the company to set, track, and manage timelines, deliverables, budgets, and scope to ensure successful program execution
• Deliver clear, timely, and consistent updates across engineering, leadership, and non–technical teams on the progress, plans, and incidents
• Act as a strong tactical and strategic partner to the senior leads across your program, whether that be providing low-level tactical support or help answer high-level strategic problems that drives impact at the company-level.

Requirements

• 5+ years of Technical Program Management experience focusing on Machine Learning Infrastructure, specifically covering areas such as model inference, serving, efficiency, and endpoints design & implementation
• In-depth technical knowledge around ML infrastructure design and implementation, as well as engineering best practices, supporting a balance of velocity and structure in rapidly growing organizations
• Experience working in a chaotic, fast paced, low structure environment. We need an EPM who is pragmatic, can roll up their sleeves when needed, and can function at the tactical and strategic level to do whatever it takes to make Cohere’s models succeed

Qualifications

• Have technical experience in a hands-on capacity. We don’t necessarily need someone who has been working hands-on in a technical role, but the programs are inherently extremely technical. You need to dive into the depths of internal systems, and develop a reputation as a subject matter expert who brings in engineering best practices and the right amount of structure to continuously improve the program
• Experience working with ML teams, and a working understanding of how machine learning models are built in order to advocate for both the needs of the infrastructure teams and modeling teams in building scalable, highly performant ML infrastructure.

Benefits

• 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.