Lyric is focused on simplifying the business of care by preventing inaccurate payments and reducing waste in the healthcare ecosystem through AI-first technology. The company aims to enable more efficient use of resources to reduce the cost of care for payers, providers, and patients.
Requirements
- Minimum of seven (7) years of experience in AI/ML engineering, with at least three (3) years in a technical or team leadership role
- Previous Technical Leadership in the AI/ML leadership space
- Hands-on experience building and deploying S/LLMs or generative AI applications (e.g., using Llama, Deepseek or similar frameworks)
- Proven track record of extracting structured data from unstructured document sources, including scanned forms, free-text reports, and complex layouts
- Strong software engineering skills in Python and ML frameworks (e.g., Kubeflow, PyTorch, multi-agentic frameworks)
- Experience with OCR technologies (e.g., Tesseract, Amazon Textract), NLP techniques, and model deployment in production environments
- Deep understanding of NLP methods including embeddings, transformers, named entity recognition (NER), and text classification
Responsibilities
- Lead the architecture, development, and deployment of AI/ML systems for document ingestion, understanding, and data extraction.
- Build and create good datasets and a system of good validation and verifications of data and ML systems.
- Build and fine-tune LLMs and generative AI models to interpret, summarize, and extract information from complex unstructured content.
- Develop NLP pipelines leveraging techniques such as OCR, entity recognition, text classification, summarization, and semantic parsing.
- Integrate LLMs with retrieval systems (RAG), vector databases, and structured outputs suitable for downstream consumption.
- Collaborate cross-functionally to align technical solutions with product requirements and compliance needs.
- Mentor a team of AI/ML engineers, establish best practices in model training, evaluation, and monitoring.
Other
- Stay abreast of the latest advancements in generative AI and apply cutting-edge techniques to real-world document challenges.
- Experience implementing retrieval-augmented generation (RAG), prompt engineering, or fine-tuning foundation models
- Familiarity with vector databases (e.g., Postgres-pg-vector, Pinecone, FAISS, Weaviate) and semantic search
- Strong experience shipping production ML systems with a track record of monitoring and improving the ML systems
- Experience working in regulated domains such as healthcare, legal, or finance
