Update README

1 files changed, 51 insertions, 2 deletions

diff --git a/README.md b/README.md
index eb12cab..5c4f217 100644
--- a/README.md
+++ b/README.md
@@ -2,9 +2,53 @@
 
 RAG with ollama and qdrant.
 
-tbd
+## Usage
 
-### TODO
+### Setup
+
+#### Environment Variables
+
+Create a .env file or set the following parameters:
+
+```.env
+CHUNK_SIZE = <CHUNK_SIZE>
+CHUNK_OVERLAP = <CHUNK_OVERLAP>
+
+ENCODER_MODEL = <ENCODER_MODEL>
+EMBEDDING_DIM = <EMBEDDING_DIM>
+
+GENERATOR_MODEL = <GENERATOR_MODEL>
+
+RAG_DB_NAME = <DOCUMENT_DB_NAME>
+RAG_DB_USER = <RAG_DB_USER>
+
+QDRANT_URL = <QDRANT_URL>
+QDRANT_COLLECTION_NAME = <QDRANT_COLLECTION_NAME>
+```
+
+### Ollama 
+
+Download the encoder and generator models with ollama:
+
+```sh
+ollama pull $GENERATOR_MODEL
+ollama pull $ENCODER_MODEL
+```
+
+### Qdrant
+
+Qdrant will is used to store the embeddings of the chunks from the documents.
+
+Download and run qdrant.
+
+### Postgres
+
+Postgres is used to save hashes of the document chunks to prevent document chunks from 
+being added to the vector db more than ones.
+
+Download and run qdrant.
+
+#### Running
 
 Build script/or FE for adding pdfs or retrieve information
 
@@ -12,8 +56,13 @@ Build script/or FE for adding pdfs or retrieve information
 
 [streamlit](https://github.com/streamlit/streamlit)
 
+### Notes
+
+Yes, it is inefficient/dumb to use ollama when you can just load the models with python
+in the same process.
 
 ### Inspiration
+
 I took some inspiration from these tutorials.
 
 [rag-openai-qdrant](https://colab.research.google.com/github/qdrant/examples/blob/master/rag-openai-qdrant/rag-openai-qdrant.ipynb)