SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Jerry is a YC-backed, profitable Series C startup with 5M+ customers building an AI-powered advisor to manage physical assets (cars, homes, motorcycles, RVs, boats). You'll own the AI systems layer of a consumer product used by millions, working with LLM-based agents and internal APIs in a $2T fragmented market.
The role focuses on the systems architecture around AI—not just prompts, but agent permissions, integration patterns, guardrails, and evaluation frameworks. Jerry's existing chatbot infrastructure handles 70% of inbound sales and service requests (50,000+ chats monthly in production). This system has outgrown ad hoc ownership and needs strategic leadership.
You will decide what to build next in AI and make the case for it. Partner with OpenAI to evaluate and deploy next-generation voice models and workflow automation. Evaluate emerging AI technology (voicebot, computer use agents) and prototype what's worth building. Design evaluation frameworks and guardrails, balancing latency, cost, and accuracy in a regulated production environment. Set the technical bar for chatbot systems, shape the prompt engineering team's work, and keep architecture sound as it scales. Build new automations across record disputes, email/callout reduction, and document validation. Work directly with subject matter experts to turn domain knowledge into systems that behave correctly.
You are a systems designer who thinks in interfaces, failure modes, and blast radius. You're fluent in modern AI—reading release notes when new models drop and building with them immediately. You have informed opinions about evaluation design from hands-on experience. You're pragmatic, knowing when to build durably versus shipping what works this quarter.
Ideal experience: 3+ years in product engineering or forward-deployed engineering at early-stage startups. Track record of turning ambiguous problems into running systems. Hands-on experience putting LLM-based systems into production, including evals, regressions, latency, cost, and failure handling.
