No Commute

About the Job:

LaunchDarkly is hiring a Full Stack Engineer for the Foundation team. Foundation owns the first mile of the developer journey - the moment a developer finds LaunchDarkly, creates an account, ships their first flag, and starts to see real value. We are responsible for turning initial interest into long-term adoption, and for the PLG metrics that track how well we do it: time to value, activation, developer-to-paid conversion, and the path from a free trial to a paying customer.

We are building the next version of this experience around agents. An onboarding agent meets a developer where they already work, reads the repo in front of them, and wires LaunchDarkly into their code in a few minutes. That shape of product requires engineers who already live in this space. We want someone who has shipped features on top of MCP servers, who is comfortable designing around model capabilities, and who has a real point of view on where AI meaningfully changes a product and where it just adds surface area.

In this role, you will partner closely with product, design, data, and other engineering teams to take initiatives from idea to shipped experiment. You will build agentic onboarding flows, extend LaunchDarkly into the tools developers already use, and move the PLG metrics that matter. You will have real latitude on stack, approach, and experiment design, and the bar is that the funnel moves.

Foundation ships often, iterates in small increments, and dogfoods LaunchDarkly's own flags and experimentation on everything it builds. If you want to reshape how developers adopt a platform, and you want to do it with agents rather than forms, this is the role.

Responsibilities:

• Drive features from problem framing through design partnership, full-stack implementation, instrumentation, rollout, and experimentation. Each initiative should have a clear line of sight to a PLG metric.

• Build the next version of LaunchDarkly's onboarding, including agentic flows that work inside a developer's IDE and plug into MCP servers and LLM-powered tooling. The goal is customers with a useful flag running in production on day one.

• Instrument and analyze the growth funnel: signup, activation, first flag, first experiment, first paid conversion. Run experiments that move each step.

• Ship MCP server capabilities and agent skills that extend LaunchDarkly into the tools customers already use.

• Partner closely with our PM (Natalie Friedman), design team, and data analyst to turn qualitative signal from self-serve customers into shippable experiments.

• Dogfood LaunchDarkly. Use flags, guarded rollouts, and experimentation on your own work. Give the product team real feedback on the developer experience.

• As part of our you-build-it-you-run-it culture, all developers may be responsible for supporting applications in production, including on-call.

Qualifications:

• 5+ years of full-stack engineering experience, with a portfolio of features you personally drove from idea to production.

• Shipped AI-first features in production. You have used MCP servers, agent frameworks, or LLM tool-use in real products, not just demos, and you have a point of view on what works and what does not.

• Comfort across the stack. Strong with TypeScript and React on the frontend. Able to hold your own in Go or Node on the backend. Our stack is TypeScript and Go.

• Track record of moving product-led growth metrics. Hands-on experience with instrumentation, funnel analysis, and A/B testing. You think in leading and lagging indicators.

• An experimentation habit. You are comfortable defining a hypothesis, wiring up a flag, running the test, reading the result honestly, and changing your mind.

• Good instincts for what makes developer onboarding actually work. You have an opinion about when friction is a bug and when it is a feature.

• A collaborative engineer who works well alongside product and design, communicates crisply, and enjoys the parts of the job that are not code.

• Genuine interest in where AI belongs in a developer platform. You have thought about where agents help in your own workflow, and you want to bring that thinking into a product that a lot of developers will touch.