About the job The Company Glassbox's mission is to reveal the insights that empower organizations to deliver exceptional digital customer experiences. Glassbox is a leading force in shaping digital experiences. It helps organizations uncover digital issues, boost conversion rates, enhance accessibility, prevent fraud, and more. Leveraging AI-driven customer intelligence, Glassbox enables enterprises to deliver secure, proactive, and preventative digital experiences. Its solutions are trusted by highly regulated organizations, including SoFi, Cal, and many others. We are growing and have been recognized by G2 as one of 2024's Top 50 Software Companies in the world. About Us We are building an AI-native R&D platform that transforms how engineering teams operate. Our vision is to move from reactive workflows to intelligence-driven systems, where AI agents are embedded across the entire software development lifecycle - from code to production. These agents will both: * Work alongside engineers as copilots * Operate independently as autonomous systems The RoleWe are looking for ahands-on AI Agent Engineerto design and build production-grade AI systems that integrate deeply into engineering workflows.This is not a research role and not a prompt-engineering role.You will build real systems that: * Are used daily by engineers * Run independently and take action * Deliver measurable impact on speed, quality, and reliability What You'll Do * Design and build AI agents that: * Assist engineers in real-time (copilot mode) * Operate autonomously and take actions (self-sufficient mode) * Develop end-to-end agent workflows across systems such as: * CI/CD pipelines * Code repositories * Observability and monitoring tools * Support and ticketing systems * Build core components: * RAG pipelines (retrieval, embeddings, vector databases) * Tool-integrated agents (APIs, services, workflows) * Agent orchestration, memory, and state management * Turn LLM capabilities into production-grade systems by: * Handling failures, retries, and edge cases * Managing latency, cost, and scalability * Implementing observability (e.g., Langfuse or similar) * Work closely with engineering teams to ensure: * Real adoption * Seamless integration into workflows * Balance between autonomy and human control Requirements * Strong backend or platform engineering experience (2–5+ years) * Proven experience building and deploying LLM-based agents in production * Hands-on experience with: * LangChain, LangGraph, or similar frameworks * LLM APIs (OpenAI, Anthropic, etc.) * RAG systems (vector DBs, embeddings, retrieval pipelines) * Tool-based agents (APIs, workflows, automation) * Experience with Claude (Anthropic), including Claude Code - mandatory * Used as part of development workflows or integrated into systems * Not limited to chat usage * Experience with observability tools such as Langfuse (or similar) * Strong coding skills in TypeScript or Python Nice to Have * Experience building multi-agent or autonomous systems * Experience designing human-in-the-loop workflows * Background in DevOps, CI/CD, QA, or DevEx platforms * Experience with event-driven systems (Kafka, queues) * Experience integrating AI into: * Code review * Testing / QA * Incident analysis Who You Are * A builder - you ship systems end-to-end * You think in systems and workflows, not just prompts * You are comfortable going from idea to production * You care about impact, reliability, and real-world usage Why Join UsYou'll be building the foundation of anAI-native engineering platform, where: * Agents are first-class components across R&D * Systems are autonomous, connected, and continuously improving * Engineering teams are augmented - not replaced - by AI This is an opportunity to shape how modern engineering organizations operate in the AI-first era.