Sift Healthcare - Senior AI Engineer

• Generative AI & LLMs, Monitoring and Observability: Utilization and deployment of LLMs for clinical decision support, and revenue-cycle automation. Implement systematic LLM evaluation and monitoring, including task-level metrics, LLM-as-a-judge scoring, and detection of input, embedding, and output drift to ensure stable production performance. • Retrieval: Build and optimize RAG pipelines, including document ingestion, chunking, embeddings, and retrieval. Implement evaluation and monitoring frameworks to assess retrieval and generation quality and detect drift in data, embeddings, and outputs. • Deep Learning: Research, develop, and optimize deep learning models for healthcare revenue-cycle applications, from experimentation and training through evaluation and assisting with deployment. Leverage embedding architectures such as TitanV2 and ClinicalBERT variants, along with other transformer-based models, to understand clinical documentation, denials, and payer communications. • Unsupervised & deep clustering: Design deep clustering pipelines, leveraging techniques such as HDBSCAN for scalable density-based clustering and IDEC (Improved Deep Embedded Clustering) for joint representation learning and clustering. • Best practices & compliance: Uphold rigorous code quality standards, conduct peer reviews, adhere to Git workflows, and ensure HIPAA-compliant handling of PII/PHI data. • Cross-functional collaboration: Partner with Data Scientists, ML & Dev Ops, Engineers, and Product teams to integrate deep learning capabilities into our platform and drive data-driven features.