Sr. Product Engineer

Product Engineer — Nolte
Nolte is a consultative product partner for companies in regulated industries (insurance, healthcare, fintech). We don't just build what clients ask for. We challenge product theses, protect founders from bad bets, and ship predictably using our own delivery framework and forecasting system (NolteOS). We're an Anthropic partner and AI-native across our workflow.

The role
You'll own the full arc from "should we build this?" to "it's live in production." That means leading discovery conversations with clients, pressure-testing product decisions before a line of code is written, scoping deliveries, building, and shipping. You're the person in the room who can explain why something shouldn't be built, and be right, then turn around and ship the thing that should.

What you'll actually do

• Lead client-facing conversations: discovery calls, sprint reviews, roadmap discussions. You're not presenting status updates. You're shaping what gets built and why.
• Evaluate product bets critically. If a feature doesn't tie to a business objective or the risk profile is wrong, you say so with evidence, not vibes.
• Scope and deliver work within Nolte's delivery model: discrete, validated units tied to business outcomes with defined acceptance criteria.
• Build across the stack in TypeScript/JavaScript. Our work lives in Node, React, Next.js, and the ecosystems around them. You're writing application code, not configuring templates.
• Write and maintain BDD test suites (Cucumber/Gherkin) aligned to our delivery methodology. Acceptance criteria aren't an afterthought; they drive implementation.
• Work with the Anthropic API, Claude Code, and MCP integrations as core parts of your development workflow. You're building with and on top of LLMs, not just prompting them in a chat window.
• Architect and implement AI-augmented features for client products: structured output pipelines, tool use patterns, retrieval systems, agent workflows. This is hands-on integration work, not experimentation.
• Manage ongoing client relationships with the attentiveness and communication quality of a great account manager. Proactive updates, clear expectations, no surprises.

What this isn't

• A pure engineering role where someone else decides what to build
• A PM role where someone else writes the code
• A junior position with a growth path into this. We need someone already operating here

Requirements

• You've been the person responsible for both the "what" and the "how." Product thinking and technical execution in the same seat.
• You're deep in JavaScript/TypeScript. You have opinions about runtime tradeoffs, bundler configs, type safety patterns, and monorepo structure, and those opinions come from building production systems, not reading blog posts.
• You've built on LLM APIs directly: function calling, streaming, structured outputs, prompt engineering at the application layer. You understand token economics, latency tradeoffs, and where to draw the line between AI and deterministic code.
• You can hold your own in a business conversation with a founder or VP Product. You understand unit economics, regulatory constraints, go-to-market timing, and why those things change what you build.
• You've said no to a stakeholder and it was the right call. You can point to a time you killed or reshaped a feature that would have been a bad bet.
• Your communication is a strength, not a tolerable weakness. Written and verbal, client-facing and internal. You write clearly, you listen well, you de-escalate naturally.
• You're comfortable with ambiguity and ownership. We don't hand you a fully scoped ticket queue. You help define the work, then do it.

What you'll actually do

• Lead client-facing conversations: discovery calls, sprint reviews, roadmap discussions. You're not presenting status updates. You're shaping what gets built and why.
• Evaluate product bets critically. If a feature doesn't tie to a business objective or the risk profile is wrong, you say so with evidence, not vibes.
• Scope and deliver work within Nolte's delivery model: discrete, validated units tied to business outcomes with defined acceptance criteria.
• Build across the stack in TypeScript/JavaScript. Our work lives in Node, React, Next.js, and the ecosystems around them. You're writing application code, not configuring templates.
• Write and maintain BDD test suites (Cucumber/Gherkin) aligned to our delivery methodology. Acceptance criteria aren't an afterthought; they drive implementation.
• Work with the Anthropic API, Claude Code, and MCP integrations as core parts of your development workflow. You're building with and on top of LLMs, not just prompting them in a chat window.
• Architect and implement AI-augmented features for client products: structured output pipelines, tool use patterns, retrieval systems, agent workflows. This is hands-on integration work, not experimentation.
• Manage ongoing client relationships with the attentiveness and communication quality of a great account manager. Proactive updates, clear expectations, no surprises.

Benefits

• 30 days paid time off
• Annual remote retreat
• Expense and equipement allowance
• Self-managed schedule
• Profit sharing