Senior Product Manager Job

Job Title: Senior Product Manager
Job Duties:
• Analyze current and anticipated customer needs, model capabilities and deployment requirements, and Cerebras’s AI inference platform capabilities to establish technical requirements, project selection criteria, implementation priorities and project sequencing for models supported on the Cerebras platform. (15% of the time)
o Review technical research papers describing model architectures and algorithms, together with model cards, configuration files, and documentation available through Hugging Face and GitHub, to identify requirements affecting model compatibility, implementation effort, and production deployment on Cerebras inference cloud.
o Collaborate with model-enablement and performance engineers to evaluate model architecture characteristics, including attention mechanisms, dense versus mixture-of-experts designs, numerical precision, and context and memory requirements, and identify platform capabilities, technical dependencies, and implementation requirements affecting project planning and deployment.
o Build and run evaluation suites comprising representative customer tasks, including coding tasks and performance benchmarks, to assess candidate models through model-provider APIs, OpenRouter, and Hugging Face resources, as applicable; document task correctness, response quality, latency, and other relevant results; and use the findings to establish project-selection criteria, technical requirements, and acceptance criteria for model-enablement projects.

• Plan, initiate, and manage cross-functional projects to enable selected models and required platform features, including defining project objectives, technical requirements, milestones, dependencies, resource needs, schedules, and launch-acceptance criteria. (20% of the time)
o Coordinate and conduct cross-vendor evaluation of AI models using software tools and platforms, such as GitHub, VS Code, Claude Code, and Codex, to assess capabilities and behavior, identify suitability and implementation considerations, and support project technology-selection decisions.
o Develop project plans for onboarding customer-provided fine-tuned models to Cerebras dedicated deployments. Coordinate model intake, architecture-compatibility assessment, required platform support, optimization, evaluation, customer acceptance, and production launch.
o Identify dependencies among model-enablement work, platform software, performance optimizations, serving interfaces, and deployment capacity. Establish deliverables and responsible teams for each stage, and use engineering assessments to determine feasible schedules, resource requirements, and project sequencing.
o Define measurable acceptance criteria with engineering and customer teams, including task-based quality thresholds, supported context lengths, required API behavior, latency, throughput, and capacity. Specify the evaluation suite, model version, deployment configuration, and workload conditions against which readiness will be assessed.

• Serve as a liaison among customers, product teams, model-enablement engineers, performance engineers, customer engineers, and GTM teams, translating customer use cases, model-architecture characteristics, and deployment constraints into actionable product roadmaps and implementation plans. (15% of the time)
o Translate customer requirements for base and fine-tuned models into technical specifications covering intended tasks, input and output characteristics, context lengths, request concurrency, response-time expectations, and required model and API capabilities.
o Facilitate technical reviews to reconcile customer requirements with model behavior, architecture compatibility, engineering feasibility, and validated platform capabilities. Document implementation alternatives, unresolved questions, and their effects on scope, resources, and launch timing.
o Maintain requirements, evaluation summaries, and decision documents that communicate agreed capabilities, technical limitations, dependencies, and acceptance criteria across teams. Incorporate engineering findings and customer evaluation results into roadmaps and implementation plans.
o Coordinate with engineers to test models on the Cerebras platform to verify compatibility, performance, and readiness against established project requirements.

• Evaluate the technical and business implications of supporting new models and platform capabilities, including architecture compatibility, inference requirements, customer demand, implementation complexity, performance expectations, capacity requirements, and operating costs; recommend IT portfolio investments and project sequencing. (15% of the time)
o Assess the feasibility of supporting candidate models and customer fine-tuned variants by reviewing architecture requirements and engineering estimates of the software support, memory, compute, and dedicated deployment capacity needed for production use.
o Evaluate baseline and proposed deployment configurations using representative coding tasks, customer workloads, and performance benchmarks to assess quality, performance, and cost tradeoffs. Recommend implementation and optimization priorities based on whether each configuration meets customer requirements and intended latency, throughput, and capacity targets, and incorporate the findings into project and deployment decisions.
o Prepare investment and sequencing recommendations explaining expected customer benefit, implementation effort, technical risks, resource requirements, and opportunities to reuse platform capabilities across model projects. Document the evaluation evidence and assumptions underlying each recommendation.

• Monitor model-enablement and launch projects against established milestones, scope, quality standards, and readiness criteria; identify risks, blockers, and resource needs; coordinate their resolution; and revise project plans as necessary. (15% of the time)
o Coordinate monitoring of model performance and activity using agentic scripts and Grafana, tracking performance issues, capacity constraints, and operational risks and coordinating follow-up actions with engineering teams to maintain deployment readiness.
o Track model intake, enablement, optimization, evaluation, and customer-validation milestones against the project plan. Review engineering progress and test results against documented technical requirements and launch-acceptance criteria.
o Maintain and execute repeatable evaluation suites to verify project deliverables before and after platform optimizations. Identify regressions in coding-task correctness, response quality, latency, throughput, or other agreed measures; coordinate investigation with engineering teams; and use the findings to revise project risks, remediation priorities, and readiness assessments.
o Assess the downstream effects of compatibility gaps, performance shortfalls, serving limitations, and capacity constraints. Update project scope, dependencies, milestones, and readiness assessments, and prepare status reports explaining affected deliverables, proposed resolutions, and impacts on customer launch commitments.

• Lead technical launch and implementation readiness for supported models by coordinating validation signoffs, support plans, customer and field enablement, release materials, and launch communications, ensuring that external commitments accurately reflect validated platform capabilities and project readiness. (10% of the time)
o Coordinate final qualification of supported models and customer fine-tuned models on the intended Cerebras deployment configuration. Consolidate task-based evaluation and performance-benchmark results, track unresolved limitations, and obtain engineering and customer-validation signoffs against the agreed launch criteria.
o Coordinate and prepare implementation and support materials identifying the validated model version, deployment configuration, supported features, input and output limits, known restrictions, and escalation procedures. Develop launch guides and other technical materials, including software parameters and code samples, and coordinate with engineering and field teams to ensure customer stakeholders understand the conditions affecting implementation and deployment.
o Review release materials and launch communications for consistency with validated results and approved project scope. Ensure performance statements identify applicable workload and deployment conditions and that availability and feature commitments match the configuration qualified for production.

• Assess customer use cases and technical requirements to guide appropriate model and deployment choices; coordinate resolution of technical escalations and customer-impacting issues; collect and analyze customer feedback; and incorporate findings into project roadmaps, implementation plans, status reports, and decision documents to support project prioritization and execution. (10% of the time)
o Assess customer workflows, including coding assistants and multi-step AI-agent applications, and translate business needs into technical requirements for task quality, model capabilities, context handling, latency, throughput, and serving-interface behavior. Coordinate development of representative evaluation tasks to assess implementations, inform model and deployment recommendations, and verify that proposed implementation meets customer requirements.
o Evaluate coding command-line tools and review agent-infrastructure repositories and documentation on GitHub to identify integration requirements and implementation considerations. Translate findings concerning tool calling, structured responses, streaming, and repeated inference requests into evaluation cases, technical requirements, and implementation priorities, where applicable.
o Collect customer issue descriptions, request examples, evaluation results, and reproduction conditions; coordinate technical investigation with customer and engineering teams; and document customer impact and resolution dependencies. Incorporate recurring issues and validated feedback into evaluation suites, acceptance criteria, project plans, and prioritization decisions.

Minimum Requirements:
Bachelor's degree or its equivalent in Software Engineering or a related field, and 3 years of related work experience.
In addition to general information-technology project-management competencies, the position requires the following AI-domain knowledge:
• Knowledge of generative AI and large-language-model architectures, including attention mechanisms, dense and mixture-of-experts models, model deployment and serving requirements, inference workload characteristics, and the open-model ecosystem.
• Knowledge of production-inference considerations for base and fine-tuned models, including model-to-platform compatibility, context and memory requirements, compute requirements, latency, throughput, deployment capacity, and quality-performance-cost tradeoffs.
• Ability to interpret technical research, model documentation, engineering assessments, and task-based evaluations and apply the findings to model-enablement requirements, optimization priorities, project sequencing, customer acceptance, and production-launch readiness.

Employer: Cerebras Systems Inc.
Salary Range: $196,000.00 per year to $215,000.00 per year
If you are interested in applying for this position, please email resume with Job # 202610 to [email protected]