USA_Developer

· Strong day-to-day experience with Claude Code, including a disciplined workflow: plan before coding, use CLAUDE.md project memory, implement in small steps with tests, and review the code it generates. · Has run parallel workstreams using git worktrees, multiple sessions, subagents with focused roles (such as tester, reviewer or refactorer), custom slash commands, hooks and MCP integrations. · Can deliver deployable progress on several projects in the same day without sitting idle while a model generates. Applied AI engineering · Has built and taken to production RAG systems, with hands-on use of parsing and chunking libraries (for example Unstructured, Docling, LangChain or LlamaIndex splitters). · Solid grasp of hybrid search (BM25 plus dense vectors, reciprocal rank fusion) and cross-encoder reranking. · Has built agentic and multi-agent systems with LangGraph or a similar framework (CrewAI, AutoGen, Semantic Kernel), and understands state, checkpointing, conditional routing and interrupts. · Has built LLM evaluation using golden datasets, retrieval metrics (recall@k, MRR), faithfulness and groundedness checks, and LLM-as-judge, with tools such as Ragas, DeepEval or Promptfoo. · Can optimize cost and latency through model routing, prompt caching, context reduction and batching. · Strong Python and FastAPI, plus working knowledge of Docker and CI/CD. Open-source stack and best practices · Embeddings: open-source models such as bge, e5 or nomic, with the ability to benchmark them against commercial options. · Knows how to avoid vendor lock-in, check licenses on models and libraries, pin versions, and keep components modular so they can be swapped. Cloud · Preferred: Azure and GCP. For example, Azure OpenAI, Azure AI Search, Azure AI Document Intelligence, Vertex AI, Vertex AI Vector Search, GKE or AKS. Nice to have · Healthcare domain experience (payer or provider), with knowledge of PHI and HIPAA requirements and familiarity with FHIR. · Guardrails and security for LLM apps: PHI masking, prompt-injection defense, grounded answers with citations, audit logging. · Fine-tuning experience (LoRA or QLoRA) on open-weight models. Roles & Responsibilities We’re looking for a hands-on Applied AI Engineer who builds and ships AI products. You’ll join a team that finds AI opportunities across the business and turns them into working POCs and production applications, including chatbots, document intelligence, and agent-driven workflows. You’ll often run 3-4 projects at once and deliver real progress on each one every day. To do that, you’ll need to use AI coding tools well. · Design, build and productionize RAG, agentic AI and multi-agent orchestrated applications, from POC through deployment and support. · Own the full pipeline: ingestion and parsing, chunking, embeddings, indexing, hybrid retrieval, reranking, generation, evaluation and monitoring. · Build multi-agent workflows with state management, tool calling, retries and human-in-the- loop checkpoints. · Put evaluation frameworks and guardrails in place for accuracy, faithfulness, safety and cost. · Use Claude Code daily to move several independent projects forward in parallel.

Experience working in agile development projects and sprint deliver. A builder, not a strategist. Candidates should be able to show code they wrote, systems they shipped and production problems they fixed. The interview includes a live hands-on exercise using Claude Code. Key Words to search in Resume AI Engineer, cloud, LLM, python, scalable solution, Azure, GCP, Claude Skills: AI and Automation Experience Required: 8-10 years Skills: Category Name Required Importance Experience SkillCategoryTest1_MN AI & Gen AI - Products & Tools Yes 1 7 years SkillCategoryTest1_MN AI Agents Yes 1 7 years