Java AI Agent Engineer

Design, build, and deploy AI agents and agentic workflows using Java 21+, Spring Boot 3.x, Spring AI, or LangChain4j. Build production APIs and backend services using Spring Web, Spring Data, and Spring Security. Implement agent capabilities including tool calling, orchestration, memory/context, multi-step workflows, retries, and guardrails. Design prompts and structured outputs and apply RAG, including embeddings, chunking, vector databases, and retrieval optimization. Design and maintain LLM evaluations (evals) to measure quality, reliability, and performance. Evaluate and select AI models based on quality, latency, cost, and risk. Optimize LLM applications for performance and cost, including caching, batching, model routing, and prompt/token optimization. Implement LLM observability and monitoring using tools such as Langfuse, Arize, Weights & Biases, Datadog, or similar to track traces, latency, token usage, cost, and failures. Use AI coding tools such as GitHub Copilot, Amazon Q, or Claude Code throughout development, testing, debugging, and refactoring. Collaborate with engineers and stakeholders to prototype, test, and continuously improve AI capabilities while maintaining clean, well-tested Java code. Required Qualifications 1. Java / Backend

Strong professional experience with Java, including modern Java versions such as Java 21+. Strong experience with Spring Boot 3.x and production APIs/backend services. Strong understanding of software engineering fundamentals, testing, debugging, and API design. 2. AI / Agent Development Hands-on experience building and deploying AI agents or agentic applications to production. Experience with Spring AI, LangChain4j, or comparable LLM/agent frameworks. Understanding of agent architecture, including tool calling, orchestration, memory, retries, and guardrails. Hands-on experience with LLM prompt design and RAG. Experience designing or developing LLM evaluations (evals). Experience with LLM observability/monitoring in a production environment. 3. AI-Assisted Development Daily hands-on experience with AI coding assistants such as GitHub Copilot, Amazon Q, or Claude Code. Experience incorporating AI into coding, testing, debugging, and development workflows. Experience with spec-driven or spec-to-code development is highly preferred. Preferred Qualifications Experience optimizing LLM applications for performance and cost. Experience benchmarking and selecting models based on quality, latency, cost, and risk. Experience with MCP, OpenAPI-based tool schemas, or enterprise API/tool integrations. Experience deploying AI services to AWS, Azure, or GCP. Experience with CLI-based coding assistants such as Copilot CLI or Claude Code.