LLM/RAG-Based Care Plan Generation Prototype Developer
Project Overview:
We are seeking a skilled developer to create a prototype for an LLM-based application that automatically generates care plan documents. The prototype will leverage a knowledge base constructed from existing documents and utilize a prompt sequence to drive the plan generation process.
Job Description:
We are looking for a Python developer with a strong understanding of Large Language Models (LLMs) to build a prototype for a care plan generation application. The ideal candidate will have experience with:
LLM Architecture: Demonstrated knowledge of LLM architecture and best practices, including RAG (Retrieval Augmented Generation), embeddings, and document processing.
Python Proficiency: Strong Python programming skills and experience with relevant libraries (e.g., LlamaIndex, Langchain).
Prototype Development: Ability to create functional prototypes using tools like Ollama or Langflow.
Knowledge Base Construction: Experience in extracting and structuring information from existing documents to create a comprehensive knowledge base.
Experience with RAG
Prompt Engineering: Skill in crafting effective prompts to guide LLM generation of care plans.
Responsibilities:
Develop a prototype using Python and selected tools (Ollama, Langflow, LlamaIndex) to generate care plans based on a given prompt and knowledge base.
Construct a knowledge base from provided documents using appropriate techniques.
Design and implement a prompt engineering strategy to optimize care plan generation.
Evaluate the prototype's performance and suggest improvements.
Deliverables:
A functional prototype capable of generating care plans based on a given prompt and knowledge base.
Detailed documentation of the prototype's architecture, code, and usage.
The prototype must run in offline mode, using open source models.
Required Skills:
Python programming
Open source LLMs (Mistral, Llama 3)
RAG, GraphRAG
Natural Language Processing (NLP)
PDF document processing in Python
Knowledge Graph or Semantic Search
Prototype development (Ollama, Langflow, or similar)
Preferred Skills:
Experience with healthcare domain
Experience working with time series
Location: Anywhere
Posted: Aug. 16, 2024, 8:11 a.m.
Apply Now Company Website