# The Product Builder Job Ladder A complete, fork-ready job-description ladder for an AI-native product organization. Four levels, one consistent shape, calibrated to scale from your first Builder PM hire to your most senior IC. Use it as-is, edit for your org's compensation bands, or strip the headers and drop into your ATS. --- ## How the Ladder Is Structured Every level uses the same six sections so candidates and hiring managers can compare apples to apples: 1. **The Short Version**, what the role is and isn't 2. **What You Will Actually Do**, the concrete responsibilities 3. **What We Are Looking For**, must-haves, strong signal, not required 4. **What You Will Not Do**, the deliberate non-responsibilities 5. **How We Work**, the operating model that makes the role possible 6. **How to Apply**, the work-sample submission The ladder dimensions that change between levels: - **Scope.** One feature → product area → pillar → company strategy - **Architecture decisions.** One AI workflow → multi-system → cross-team primitives → company standards - **Mentorship.** None expected → reviews peers → sets the bar for Staff/Senior → mentors Staff and shapes Principal hiring - **External voice.** None required → optional → encouraged → expected - **Years.** 4+ → 6+ → 9+ → 12+ The dimensions that stay constant at every level: - You build. The day you stop shipping prototypes is the day the role ends. - You design AI systems end-to-end. - You own evals. - You think in cost and latency. - You validate with users, not stakeholders. --- # 1. Product Builder **Level:** IC · L4 equivalent **Reports to:** Senior Product Builder or Head of Product **Years:** 4+ in product, 2+ on AI products ### The Short Version This is not a traditional PM role. We are not looking for someone who writes requirements and waits for engineering. We are looking for someone who builds. You will take a workflow from concept to working prototype before a sprint starts. You will design AI systems, not just describe them. You will validate with real users, not decks. And you will own the full arc from insight to production for one product area. ### What You Will Actually Do - Prototype before planning. First output is a working prototype, not a spec. - Design AI workflows end-to-end: input, reasoning, output, action. - Own the evaluation layer for your area. - Reason about cost and latency as product decisions. - Validate with users early and often. - Ship. Fast. ### What We Are Looking For **Must-haves:** 4+ years PM, 2+ years on AI products. Demonstrated prototyping with Claude Code, Cursor, v0 or equivalent. Hands-on experience with agentic workflows. Clear understanding of RAG, prompt architecture, and when to use models versus rules. Experience defining and running evals for AI features in production. Track record of shipping. **Strong signal:** You built something on your own with AI tools. You can walk through cost and latency tradeoffs. You worked in a paired product-engineering model. **Not required:** CS degree, production-ready backend code, large-company tenure. ### How to Apply 1. One prototype or working artifact you built using AI tools. 2. Two to three sentences on the workflow it automated. 3. The cost or latency tradeoff you had to navigate. --- # 2. Senior Product Builder **Level:** Senior IC · L5 equivalent **Reports to:** Staff Product Builder or Director of Product **Years:** 6+ in product, 3+ on AI products ### The Short Version We are looking for a Builder PM who is already operating at the upgraded version of the role and is ready to raise the bar across an entire product area or set of related areas. ### What You Will Actually Do - Own a meaningful product area end-to-end. - Lead multi-system AI workflows. Decide where memory belongs, where to break a chain, where deterministic logic is safer. - Raise the prototyping bar. New PMs learn by reading what you ship. - Own production evals for your area. Catch drift before users do. - Make defensible cost and latency calls with real numbers. - Mentor Product Builders. ### What We Are Looking For **Must-haves:** 6+ years PM, 3+ on AI products. Portfolio of shipped AI features still in production. Demonstrated multi-step agentic system design. Deep familiarity with prompt architecture, RAG, evaluation design. Track record in a paired product-engineering model. Mentorship instinct. **Strong signal:** Built or maintained an eval harness. Made and defended a model-choice that affected unit economics. Ran a discovery sprint that produced a working prototype in one week. ### How to Apply 1. One AI feature you shipped that's still in production. 2. The eval or feedback loop you built around it. 3. The cost-or-latency call that changed the architecture. --- # 3. Staff Product Builder **Level:** Staff IC · L6 equivalent **Reports to:** Principal Product Builder, VP Product, or CPO **Years:** 9+ in product, 4+ on AI products ### The Short Version Staff Builders set the technical and product bar for how the company ships AI products. You will own a pillar of the product, shape the cross-team architecture of our AI systems, and define the practices around prototyping, evaluation, and observability that every other Builder PM operates within. ### What You Will Actually Do - Own a product pillar. Multiple areas, multiple Builder PMs feeding in. - Design the AI system architecture across the pillar. - Set evaluation and observability standards. Answer "is our AI getting better or worse this month?" with data. - Ship the flagship systems yourself. You are a builder, not an overseer. - Shape unit economics: cost per task, cost per user, latency budgets, pricing curves. - Mentor Senior and Product Builders. Run deep reviews. - Influence the roadmap with a strong, defensible point of view. ### What We Are Looking For **Must-haves:** 9+ years PM, 4+ on AI products. Portfolio of production AI systems with clear scale, reliability, and unit-economics evidence. Deep expertise in agentic architectures, retrieval, prompt ops, evals, drift monitoring. Lead-through-craft experience. **Strong signal:** Built an internal platform other product teams use. Owned a model migration or major prompt refactor. Public output (writing, talks, open source). ### How to Apply 1. One production AI system you designed and shipped, with architecture, scale, cost and latency profile. 2. The eval and observability layer you built around it. 3. One cross-team call you made that other people were building against six months later. --- # 4. Principal Product Builder **Level:** Principal IC · L7 equivalent **Reports to:** Chief Product Officer or CEO **Years:** 12+ in product, 5+ on AI products ### The Short Version The most senior individual contributor role in the product organization. You will set the product strategy for a flagship pillar. You will define the company-wide standards for shipping AI products. You will ship the systems the rest of the org is measured against. And you will be the most credible product voice in the room with engineering, design, and the executive team. ### What You Will Actually Do - Set the strategy for a flagship pillar. The CPO and CEO build company commitments on top of it. - Define the AI product standards for the company: prototyping, evals, observability, deprecation, model migrations, agent-fleet architecture. - Ship the flagship systems yourself. - Mentor Staff and Senior Builders. Run deep design reviews. - Influence the company alongside CTO, CEO, design leadership, and finance. - Be a public voice. Shape how the market thinks about AI products in the category. ### What We Are Looking For **Must-haves:** 12+ years in product, 5+ on AI products. Flagship AI systems led zero-to-scale. Standards adopted by other teams. Most-credible-product-voice operating range. Strategic calls that survive model generations and market shifts. Operating range from CEO whiteboard to Cursor prototype before lunch. **Strong signal:** Public output people in the field can name. Org-wide migration or repositioning. Hired and grew Staff and Senior Builders. A defensible, opinionated point of view refined over cycles. ### How to Apply 1. One flagship AI system you led from zero to scale. 2. One company-level standard you set that other teams adopted. 3. Your point of view on where AI products are heading in our category in the next 24 months. --- ## Compensation Guidance Adjust to your geography and stage. The relative shape across the ladder is what matters. | Level | Base band | Equity weight | Scope | |------|-----------|---------------|-------| | Product Builder | Mid-IC | 1x | Feature / area | | Senior Product Builder | Senior IC | 1.5–2x | Area / cluster | | Staff Product Builder | Staff IC | 2.5–4x | Pillar | | Principal Product Builder | Principal IC | 4–6x | Company strategy | --- ## Operating Model Required for This Ladder to Work This ladder assumes the operating model described in *The PM Role Is Being Rewritten in Real Time*: - Traditional PRDs are eliminated. - Product and engineering work in paired ownership. - Prototypes are expected within hours of an idea. - Reviews are outcome-based, not output-based. - Evals are launch gates, not afterthoughts. - Cost and latency are first-class product concerns. If your org still runs a write-and-wait model, the ladder will not produce the people it describes. Fix the operating model first, then hire against this ladder.