AI Engineer (42165)

I am looking for an experienced AI Engineer with strong Python and RAG/LLM expertise to join a fast-moving project focused on building an internal platform and demo toolchain. You will design end-to-end pipelines, work with hybrid retrieval, and deliver structured outputs for non-technical users. The role requires hands-on experience with FastAPI, async processing, and modern LLM workflows, along with the ability to collaborate across teams and iterate quickly in a dynamic environment.

🚀 Project
- building and iterating on an internal platform and demo toolchain focused on RAG/LLM solutions
- developing python services using FastAPI and asynchronous processing patterns
- implementing hybrid retrieval combining vector search and BM25 using Azure AI Search and/or Qdrant
- designing document ingestion pipelines including PDF extraction, OCR, chunking and metadata filtering
- generating structured outputs such as Excel and Word documents for non-technical stakeholders
- maintaining and integrating Node.js and React modules connecting frontend with Python APIs
- developing end-to-end RAG pipelines from ingestion to structured output generation
- implementing event-driven workflows using messaging patterns and tools like RabbitMQ or Azure Service Bus
- building internal demo tools using Streamlit with multi-page apps and interactive components
- ensuring engineering quality through testing, Dockerization, reproducibility and CI/CD practices

🎯 Skills
- Python 3.12+ with strong OOP and clean architecture principles
- experience with FastAPI and backend API design
- asynchronous processing using Celery and messaging/event-driven systems
- experience with PostgreSQL and relational database design
- knowledge of RAG architecture including chunking, embeddings, retrieval and generation
- prompt engineering with versioning and structured output parsing
- experience with OpenAI API and embedding models
- hybrid retrieval approach combining vector search and BM25
- experience with Qdrant or similar vector databases including metadata filtering
- integration with Azure AI Search / Azure Cognitive Search
- PDF extraction using PyMuPDF (fitz)
- OCR processing using Tesseract/pytesseract
- advanced chunking strategies for legal or financial documents
- experience with Streamlit including session state and interactive components
- data processing using pandas, NumPy and Excel generation with openpyxl
- basic to intermediate knowledge of Node.js and React for integration tasks
- experience with Docker and Docker Compose
- CLI scripting and orchestration using argparse and bash
- strong communication skills and ability to collaborate with non-technical stakeholders
- ability to debug end-to-end pipelines independently

💡 Nice to have
- experience with RAG evaluation frameworks and quality measurement
- knowledge of MCP or FastMCP
- experience with LangChain, LangGraph or multi-agent frameworks such as CrewAI, AutoGen, TaskWeave
- experience with Azure CI/CD pipelines and cloud-native deployments

#LI-MD20