LLM Engineer

Job Summary:

LLM Engineer

Tech Stack:

• Languages

  : Python (FastAPI), SQL

• LLM Orchestration

  : LangChain, custom RAG modules

• LLM Hosting

  : vLLM, MCP servers (Model Context Protocol)

• Document Processing

  : Custom chunking, PDF/Word parsers

• Vector Stores

  : FAISS, OpenSearch, Qdrant, Chroma

• Databases

  : PostgreSQL / MySQL

• Cloud

  : AWS (ECS, S3, Lambda optional)

• Containerization

  : Docker

• Task Queues

  : Celery (optional)

• Nice to have

  :

  Databricks

   for data enrichment & batch document analytics

Good to Have:

• Experience with

  Databricks

   for scalable pre-processing or analytics
• Frontend integration with React or Streamlit
• Hands-on with

  LLM-based automation

   and document understanding (OCR, ICD/CPT extraction)

Key Responsibilities:

• Design and implement end-to-end

  RAG pipelines

   using LangChain or custom orchestration.
• Build RESTful APIs using

  FastAPI

   to expose LLM-powered features.
• Develop chunking, embedding, and retrieval logic for clinical documents.
• Deploy services on

  AWS ECS

   using

  Docker

   with robust CI/CD.
• Integrate and query vector stores (FAISS, OpenSearch, or Chroma).
• Optimize

  vLLM / MCP server

   deployments for latency and throughput.
• Automate repetitive healthcare workflows using LLMs.
• Create and manage

  SQL-based databases

   and schemas.
• Monitor and log model inference calls, context size, and token cost.
• Implement authentication, rate limiting, and auditing for production APIs.
• Collaborate with data scientists, domain experts, and DevOps