FIDES is seeking an experienced AI Solutions Architect/Engineer to support a leading health system client. This role focuses on designing and implementing secure, scalable AI applications using large language models (LLMs), modern data pipelines, and advanced orchestration tools. The ideal candidate brings deep technical expertise and thrives in collaborative, agile environments where ambiguity is embraced, and systems are built to last.

Key Responsibilities:

• Lead the development and deployment of enterprise-grade AI applications leveraging commercial and open-source LLMs (e.g., OpenAI, Anthropic, Mistral, LLaMA, Mixtral).
• Fine-tune models using LoRA, PEFT, and evaluate model suitability for specific enterprise use cases.
• Build and manage semantic search and retrieval pipelines (RAG), including document chunking and vector search using FAISS, Weaviate, PGVector, or Chroma.
• Utilize agentic AI frameworks such as LangChain, CrewAI, AutoGen, or Semantic Kernel to develop robust, multi-step workflows.
• Design and orchestrate pipelines for structured and unstructured data ingestion (e.g., CSVs, PDFs, EHRs, APIs, logs) using tools like Airflow, Prefect, or Dagster.
• Apply data wrangling frameworks (Pandas, Dask, Spark) and implement entity resolution and cross-source relationship mapping.
• Build secure, scalable APIs and services using Python (FastAPI), Node.js/TypeScript, and optionally React for front-end integration.
• Integrate AI-powered tools into enterprise ecosystems such as Snowflake, ServiceNow, SharePoint, and Elastic.
• Ensure security, compliance, and scalability aligned with PHI, PII, and HIPAA standards, including RBAC, API protections, and audit logging.
• Deploy applications using Docker/Kubernetes or serverless functions; maintain CI/CD pipelines and version AI models as artifacts.
• Operate within preferred cloud stacks (Azure OpenAI, Vertex AI) and collaborate across engineering, data, and security teams.

What We're Looking For:

• 5+ years of experience designing and delivering AI/ML applications in production environments.
• Strong knowledge of prompt engineering, hallucination mitigation, and LLM evaluation strategies.
• Proven ability to architect and operationalize AI systems in ambiguous environments with strong communication and documentation skills.
• Experience building modular, reusable components and delivering user-centric tools for data interaction and insight generation.
• Familiarity with MLOps platforms such as MLflow, Ray, or Nvidia Triton.
• Comfort using GitHub Copilot, code interpreters, and other developer productivity tools.

Preferred Qualifications:

• UI/UX experience for data discovery platforms or voice/chat-based interfaces.
• Exposure to multi-modal or agentic AI, chatbot design, or voice interface applications.
• Experience working in regulated environments with a focus on compliance and scalability.
• Experience working with Azure, AWS, and Databricks.