Senior Product Manager, Experimentation Tooling

About the role

We’re looking for a Senior Product Manager to own Lightning AI’s experimentation and post-training product end to end—from product strategy and roadmap through launch, adoption, pricing, and go-to-market. This is a role focused on how AI researchers and engineers turn an idea into a high-quality, validated model.

Responsibilities

• Own the product vision and roadmap for post-training and experimentation — what we build, what we integrate with, what we don't build, and in what order
• Understand how ML engineers and AI researchers actually work today: the jobs they run, the comparisons they make, the failures they debug, and the handoffs that break down between research and production — then build the product that makes that workflow coherent
• Develop a strong point of view on where Lightning should build differentiated experiences versus integrate with the existing ecosystem of experiment trackers, evaluation frameworks, data tools, and model registries
• Work directly with engineers from problem definition through architecture, implementation, and launch — understand the constraints, help shape the solutions, don't hand off requirements and wait
• Own model evaluation as a product function — write evals, assess outputs, and let quality signals drive roadmap decisions
• Design pricing and packaging in partnership with Growth and Finance — model unit economics, run experiments, and make calls that affect both adoption and margin
• Build workflows that help teams collaborate: share results, compare models, move work from research into production, and maintain enough lineage that decisions can be explained and reproduced
• Define and instrument the metrics that matter across activation, iteration speed, compute consumption, retention, and expansion

Requirements

• 7+ years of product management experience, including at least 3 years building infrastructure, platform, developer-tooling, or machine-learning products
• Hands-on experience building products for ML engineers, AI researchers, or data scientists
• A detailed understanding of experimentation and post-training workflows, including training jobs, checkpoints, metrics, artifacts, experiment comparison, reproducibility, and model evaluation
• Experience with one or more modern post-training techniques, such as supervised fine-tuning, preference optimization, reinforcement learning, distributed training, or hyperparameter optimization
• Experience designing or working closely with model evaluations. You understand that model quality is multidimensional and know how qualitative and quantitative signals should inform product decisions
• Strong product and interaction judgment. You can simplify technically complex workflows, develop a clear information architecture, and create usable product experiences without depending heavily on a dedicated design function
• Enough technical depth to work directly with infrastructure and ML engineers on APIs, SDKs, execution systems, distributed workloads, observability, data and artifact management, and failure handling
• Record of end-to-end ownership. You do not stop at roadmap definition; you investigate problems personally, create the necessary artifacts, drive decisions, support implementation, validate the result, and help bring the product to market
• Strong prioritization and willingness to say no. You can identify the narrowest valuable product wedge and resist building a broad collection of loosely connected features
• Experience owning pricing, packaging, or consumption-based products, including working with unit economics and making decisions that affect adoption and margin
• Strong written and verbal communication. You are equally comfortable writing a technical product specification, reviewing a developer workflow, presenting to executives, and explaining the product to a customer
• High bias for action and comfort operating in ambiguous, fast-moving environments with limited process and support
• BS in Computer Science, Engineering, or equivalent practical experience

Qualifications

• Experience at an AI infrastructure company, neocloud, hyperscaler ML platform, experiment-tracking company, or developer-tooling startup
• Familiarity with PyTorch, PyTorch Lightning, distributed training, GPU infrastructure, or large-scale fine-tuning
• Experience with tools such as Weights & Biases, MLflow, Hugging Face, Ray, Slurm, Kubernetes, or comparable internal ML platforms
• Experience building collaborative workflows for teams moving models from research through evaluation and production