datadog
Product Manager II - Model Lab
At a Glance
- Location
- New York, United States
- Experience
- 4+ years
- Posted
- 2026-03-10T10:10:24-04:00
Key Requirements
Required Skills
Domain Knowledge
- Engineering
Requirements
You have 4+ years of Product Management experience, with at least one 0→1 product or major feature launch
You have experience with developer tools, ML infrastructure, data platforms, or AI/LLM systems
You understand (or can quickly ramp on) modern ML training workflows — distributed training, experiment tracking, hyperparameter tuning, artifact storage, evaluation pipelines
You have familiarity with frameworks such as PyTorch, TensorFlow, JAX, or distributed training frameworks
You can translate highly technical concepts into compelling product narratives
You thrive in cross-functional environments and can align engineering, design, and GTM around a clear strategy
Compensation & Benefits
New hire stock equity (RSUs) and employee stock purchase plan (ESPP)
Continuous professional development, product training, and career pathing
Intradepartmental mentor and buddy program
An inclusive company culture, ability to join our Community Guilds (Datadog employee resource groups)
Access to Inclusion Talks, our Internal panel discussions
Global mental health benefits for employees and dependents age 6+
Responsibilities
Define the vision and strategy for Model Lab, establishing Datadog’s position in experiment tracking and model training observability
Lead 0→1 product discovery with AI research teams, ML platform engineers, and infrastructure leaders to deeply understand experiment tracking workflows and pain points
Design a system that unifies training metrics, hyperparameters, artifacts, dataset lineage, and model evaluation into a coherent and scalable experience
Identify differentiation opportunities vs.
competitive alternatives and homegrown internal tooling
Partner closely with engineering and design to ship foundational capabilities such as experiment lineage, artifact versioning, distributed training visibility, and reproducibility workflows