SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Keycard is building identity and access infrastructure for the agent-native era—a control plane for agents that handles identity, credential exchange, delegation, policy, and audit. You'll own the evaluation journey end-to-end, transforming what field engineers currently do manually into a self-guided product experience.
The core challenge: every customer starts with an evaluation, and today a field engineer runs each one by hand. Your job is to turn that walkthrough into product so evaluations can succeed without manual intervention. This means owning the developer experience across the SDK, CLI, console, and docs—with a specific focus on reducing time-to-first-credential-exchange.
You'll define what a successful evaluation looks like for an agent-identity product, instrument the journey to identify where evaluators stall, and work with engineering to fix the worst friction points. Often the fix isn't a feature—it's a better quickstart, a template repo, clearer error messages, or docs that don't assume too much context.
Key responsibilities include: owning evaluation milestones and metrics that product, engineering, and go-to-market report against; driving developer experience improvements across all surfaces; turning field-engineering playbooks into self-guided paths; running experiments to validate improvements; incorporating evaluator feedback into the roadmap; and collaborating with DevRel, marketing, sales, and field engineering on evaluation motion.
You should have PM experience on developer products where evaluation and adoption were core metrics (dev tools, infrastructure, APIs, or security). You're technical enough to read SDK code, follow API design discussions, and pull your own analytics. You have strong taste in developer experience—you notice why one CLI feels good and another doesn't, and you care about error messages, defaults, and docs structure. You build with AI tools yourself and follow the agent ecosystem closely. You start from instrumentation and don't form theories without data. You're comfortable defining a category with no playbook. You write clearly, as much of this role's output is docs and problem statements others will act on.
Strong candidates may have engineering backgrounds, experience productizing high-touch motions, familiarity with identity/auth concepts or agent protocols, growth tooling experience, or shipped docs and quickstarts developers actually used.
