Senior Product Manager, Inference

Lightning AI•San Francisco, NY

4d•Hybrid

About The Position

We're looking for a Founding Product Manager for Inference who will own this product end-to-end—roadmap, pricing, and GTM—from the ground up. This is a zero-to-one role at the intersection of deep technical fluency and commercial instinct. You'll define what we build and why, design the developer journey from first API call to production workload, and be the product voice in sales cycles and the market. The right candidate has lived inside the machine; they've operated model serving infrastructure or shipped products on top of it, and can move fluidly between a latency/throughput tradeoff conversation with an infra engineer and a positioning conversation with a sales lead. You will be joining the Product Team and report to our VP of Product working directly with our executive team as we grow this business.

Requirements

7+ years of product management experience, with at least 3 years in infrastructure, platform, or developer tooling products
Direct, hands-on experience with model serving or inference infrastructure — you've shipped in this space; you understand quantization, batching strategies, KV cache, and speculative decoding at a level that lets you go deep with ML engineers
Proven track record owning product pricing and packaging decisions, not just feature decisions — you've modeled unit economics and made calls that affected margin
Experience with a PLG or trial-to-paid motion in a developer product; you know how to build self-serve growth loops and run rigorous A/B experiments
Strong analytical skills — comfortable with product instrumentation, metrics, and dashboards; you pull your own data
Excellent written and verbal communication; you can write a crisp one-pager, a technical spec, and a customer-facing benchmark brief with equal fluency
Bias for action and comfort operating with high ambiguity in a fast-moving environment
Bachelor's degree in Computer Science, Engineering, or related technical field (or equivalent practical experience)

Nice To Haves

Prior experience at a neocloud, hyperscaler inference team, or AI infrastructure startup
Familiarity with the PyTorch/Lightning ecosystem
Background in GPU cluster products or consumption-based infrastructure pricing

Responsibilities

Define Lightning AI's inference product vision and roadmap — what we build, what we don't, and in what order — translating the competitive landscape (vLLM, Together, Fireworks, Modal, hyperscaler inference APIs) into a differentiated strategy grounded in Lightning's compute and software advantage
Own inference pricing and packaging end-to-end: design the model (per-token, per-second, reserved capacity), run pricing experiments with Growth and Finance, and define the tiers that convert self-serve developers into enterprise contracts
Be the product voice in GTM: develop sales positioning, answer technical objections in the field, and partner with Marketing on the benchmarks, reference architectures, and developer content that builds credibility with ML engineers and platform teams
Own the developer journey from API key to production-scale deployment — identify and remove friction across onboarding, documentation, SDK ergonomics, and dashboard observability
Lead experiments across activation flows, pricing pages, and upgrade prompts; track and move DAU/MAU, Time to Value, Activation %, PQLs, and expansion revenue
Partner with engineering to write tight specs and make fast build/buy/partner decisions; collaborate across Product to ensure inference coheres with training, fine-tuning, and storage surfaces
Establish inference-specific metrics — throughput, latency SLAs, cold-start behavior, cost per token — and build the instrumentation to track them