First draft complete

Author: ajosh0504

docs/10-mongodb-atlas/2-create-cluster.mdx

@@ -55,6 +55,6 @@ If you don't see a button to **Allow Access from Anywhere**, you should close th
 
 ## That's all!
 
-That's all! You have a new database cluster. If everything goes well, you should see your newly created cluster in the database deployment screen.
+That's all! You have a new database cluster. If everything goes well, you should see your newly created cluster on the **Database** tab under the **Deployment** section.
 
 <Screenshot url="https://cloud.mongodb.com" src="img/screenshots/10-mongodb-atlas/2-create-cluster/3-database-deployment.png" alt="Screenshot of the database deployment screen" />

docs/10-mongodb-atlas/3-get-connection-string.mdx

@@ -5,7 +5,7 @@ import Screenshot from "@site/src/components/Screenshot";
 
 In order to ingest data into your cluster later in the lab, you will need to get the connection string for your cluster.
 
-In the Atlas UI, navigate to the **Overview** page. In the **Clusters section**, selct the tab for the cluster you just created and click **Connect**. 
+In the Atlas UI, navigate to the **Overview** page. In the **Clusters section**, select the cluster you just created and click **Connect**. 
 
 <Screenshot url="https://cloud.mongodb.com" src="img/screenshots/10-mongodb-atlas/3-get-conn-string/1-connect-button.png" alt="Screenshot of the connect button" />
 

docs/20-fireworks-ai/1-create-account.mdx

@@ -2,7 +2,7 @@ import Screenshot from "@site/src/components/Screenshot";
 
 # 👐 Create an account
 
-In this lab, we will be using _FireFunction V1_, a free and open-source function calling model from Fireworks AI.
+In this lab, we will use _FireFunction V1_, a free and open-source function calling model from Fireworks AI.
 
 The easiest way to use this model is via the Fireworks API. But first, you will need to create a Fireworks account. **If you already have an account, you can move on to the next section.**
 

docs/30-dev-env/1-dev-setup.mdx

@@ -1,6 +1,6 @@
 import Screenshot from "@site/src/components/Screenshot";
 
-# 👐 Setup Dev Environment
+# 👐 Setup dev environment
 
 * Clone the [GitHub repo](https://github.com/mongodb-developer/ai-agents-lab-notebooks.git) for this lab by executing the following command from the terminal:
 
@@ -30,12 +30,4 @@ jupyter notebook
 
 * In the browser tab that pops up, open the notebook named `notebook_template.ipynb`. This is the notebook you will be working in for this lab.
 
-<Screenshot url="localhost:8888/tree" src="img/screenshots/30-dev-env/1-jupyter-notebook.png" alt="Jupyter Notebook" />
-
-* Run the code cell under `Step 1: Install Libraries` in the notebook to install all the Python libraries required for the lab.
-
-<Screenshot url="http://localhost:8888/notebooks/notebook_template.ipynb" src="img/screenshots/30-dev-env/2-install-libraries.png" alt="Install libraries" />
-
-* Under `Step 2: Setup Pre-requisites`, replace `<CODE_BLOCK_1>` with the MongoDB connection string, `<CODE_BLOCK_2>` with the Fireworks API key and run the cells.
-
-<Screenshot url="http://localhost:8888/notebooks/notebook_template.ipynb" src="img/screenshots/30-dev-env/3-setup-prereqs.png" alt="Setup prereqs" />
+<Screenshot url="localhost:8888/tree" src="img/screenshots/30-dev-env/1-jupyter-notebook.png" alt="Jupyter Notebook" />

docs/30-dev-env/2-setup-pre-reqs.mdx

@@ -0,0 +1,3 @@
+# 👐 Setup pre-requisites
+
+Fill in any `<CODE_BLOCK>`s and run the cells under the **Step 1: Install libraries** and **Step 2: Setup pre-requisites** sections in the notebook.

docs/40-agent-tools/1-knowledge-base.mdx

@@ -1,22 +1,12 @@
 # 👐 Create the agent's knowledge base
 
-One of the tools that the AI research agent has access to is a question-answering tool for the agent to answer questions based on information in its knowledge base. But first, we need to create the knowledge base.
+One of the tools that the AI research agent has access to is a question-answering tool that retrieves information from a knowledge base (MongoDB collection) and uses it to answer questions. But first, we need to create the knowledge base.
 
-## Download the dataset from Hugging Face
+Fill in any `<CODE_BLOCK>`s and run the cells under the **Step 3: Create a knowledge base** section in the notebook to download a dataset of Arxiv papers from [Hugging Face](https://huggingface.co/datasets/mongodb-eai/arxiv-embeddings) and ingest it into a MongoDB collection which will serve as our agent's knowledge base.
 
-We will use a dataset consisting of a subset of Arxiv papers, to build our knowledge base. The data is available on [Hugging Face](https://huggingface.co/datasets/mongodb-eai/arxiv-embeddings).
+The answers for code blocks in this section are as follows:
 
-👐 Run the first **TWO** cells under "Step 3: Ingest Data into MongoDB Atlas" to download the dataset from Hugging Face and preview it.
-
-Notice that the dataset already has an `embedding` column, which consists of embeddings of the paper abstracts.
-
-## Ingest data into MongoDB Atlas
-
-Let's use the **PyMongo** Python library to ingest data into MongoDB Atlas. Fill in `<CODE_BLOCK_3>` to `<CODE_BLOCK_6>` and run the cells containing these code blocks.
-
-👐 **CODE_BLOCK_3**: Initialize a MongoDB Python client
-
-📚 https://pymongo.readthedocs.io/en/stable/api/pymongo/mongo_client.html
+**CODE_BLOCK_3**
 
 <details>
 <summary>Answer</summary>
@@ -27,9 +17,7 @@ MongoClient(MONGODB_URI)
 </div>
 </details>
 
-👐 **CODE_BLOCK_4**: Connect to the `COLLECTION_NAME` collection using the MongoDB client
-
-📚 https://pymongo.readthedocs.io/en/stable/tutorial.html#getting-a-collection
+**CODE_BLOCK_4**
 
 <details>
 <summary>Answer</summary>
@@ -40,9 +28,7 @@ client[DB_NAME][COLLECTION_NAME]
 </div>
 </details>
 
-👐 **CODE_BLOCK_5**: Bulk delete all existing records from the `collection` collection -- should be a one-liner
-
-📚 https://pymongo.readthedocs.io/en/stable/api/pymongo/collection.html
+**CODE_BLOCK_5**
 
 <details>
 <summary>Answer</summary>
@@ -53,9 +39,7 @@ collection.delete_many({})
 </div>
 </details>
 
-👐 **CODE_BLOCK_6**: Bulk insert `records` into the `collection` collection -- should be a one-liner
-
-📚 https://pymongo.readthedocs.io/en/stable/api/pymongo/collection.html
+**CODE_BLOCK_6**
 
 <details>
 <summary>Answer</summary>

docs/40-agent-tools/2-chat-completion-llm.mdx

@@ -0,0 +1,35 @@
+# 👐 Create a vector search index
+
+To retrieve documents from the knowledge base using vector search, you must configure a vector search index on the knowledge base collection.
+
+To do this, open the **Database Deployments** page in the Atlas UI and select **Create Index** in the lower right corner under Atlas Search.
+
+<Screenshot url="https://cloud.mongodb.com" src="img/screenshots/40-agent-tools/1-create-index.png" alt="Select create index" />
+
+Click the **Create Search Index** button.
+
+<Screenshot url="https://cloud.mongodb.com" src="img/screenshots/40-agent-tools/2-create-search-index.png" alt="Create search index" />
+
+Click **JSON Editor** under Atlas Vector Search to create your index
+
+<Screenshot url="https://cloud.mongodb.com" src="img/screenshots/40-agent-tools/3-json-editor.png" alt="The 'Create Index' page with the 'JSON Editor' tab highlighted" />
+
+
+Select the `mongodb_agents_lab` database and the `knowledge` collection, change the index name to `vector_index`, and add the following index definition in the JSON editor:
+
+```python
+{
+  "fields": [
+    {
+      "type": "vector",
+      "path": "embedding",
+      "numDimensions": 1024,
+      "similarity": "cosine"
+    }
+  ]
+}
+```
+
+:::info
+The number of dimensions in the index definition is 1024 since the Arxiv dataset we used to create the knowledge base uses Mixedbread AI's open-source [mxbai-embed-large-v1](https://huggingface.co/mixedbread-ai/mxbai-embed-large-v1) model for embeddings.
+:::

docs/40-agent-tools/2-create-vector-search-index.mdx

@@ -0,0 +1,22 @@
+# 👐 Instantiate chat completion LLM
+
+Let's instantiate the chat completion LLM to use as the "brain" of our agent and for any of the tools if required.
+
+We will use Fireworks AI's free AND open-source _firefunction-v1_ model via the _ChatFireworks_ API in LangChain.
+
+Fill in any `<CODE_BLOCK>`s and run the cells under the **Step 5: Instantiate chat completion LLM** section in the notebook to create an instance of _ChatFireworks_ with the _firefunction-v1_ model.
+
+The answers for code blocks in this section are as follows:
+
+**CODE_BLOCK_7**
+
+<details>
+<summary>Answer</summary>
+<div>
+```python
+ChatFireworks(
+    model="accounts/fireworks/models/firefunction-v1", temperature=0.0, max_tokens=1024
+)
+```
+</div>
+</details>