Background Information

Project overview

Setup

Introduction

Prompt template

LCEL

Introduction to prompt engineering

Prompts

Memory

Conversation memory

Supabase

Restoring memory

Vectorstores

Supabase vector

Ingest

Retrieval

Generation

Retrieval Augmented Generation

Output Parsing

Tools

Agents

Langsmith

Farewell

( the primaryDescriptionMarkdown is long - several lines, include images, as long as you want )

In this course we <headline of what we do>. <why this technology is good> but <here's how learning it is problem>. 

In <the length of these lessons> we cover <three salient points about what we cover. Your BEST points>. Its <however long it is> and comes with <these bonus features, and reasons to buy. code, etc.>

It's taught by <author name> who <credibility, reason you should listen to them>. <social proof, if you have it yet>

<then you can list other supporting stuff you want to add, anything you want here>

-------
Example:
-------

We created a free course to create your first GraphQL server (from scratch) using Node and TypeScript. TypeScript and GraphQL are a powerful combo, but there's a lot of confusing edges if you're getting it setup the first time.

In this course we cover the basics of getting TypeScript setup with Node, how GraphQL works, and how to set up a GraphQL server using Apollo server. It's two hours long, it comes with complete code examples, and we walk through the code step-by-step.

It's taught by my friend, Shopify-engineer Hassan Djirdeh and he's integrated what he's learned using GraphQL in production as an engineer at Shopify. We've already had over 1,300 students enroll and it's been getting great reviews so far.

// ... and so on

In this course we <headline of what we do>. <why this technology is good> but <here's how learning it is problem>. 

( the summary is short - one or two lines )

Langchain.js Bootcamp

"Conversation Memory" is like your saved text messages with a friend. Chat Assistants are often trained to consider past messages to help generate meaningful responses. In fact it may help to glance at the 

 to understand what a model sees during training.

History

You can easily verify conversational memory by introducing yourself with your name and then asking the model what your name is.

The good news for us is that the ai sdk already comes with a history solution on the clientside. The entire history is sent to our defined endpoint in 

  const currentMessageContent = messages[messages.length - 1].content;

 for now because we need to define how they interact with the prompt.

Langchain offers support for chat models. They provide a 

 for use with typed messages and a placeholder for memory.

import { ChatPromptTemplate, MessagesPlaceholder } from "langchain/prompts";
import { AIMessage, HumanMessage, SystemMessage } from "langchain/schema";

We can define our Prompt Template like this.

---
code: https://gitlab.com/fullstackio/books/newline-course-apps/jeffery-wang-langchain.js-bootcamp-app/-/tree/main/Code/module_03/lesson_03.02/protected/src?ref_type=heads
---

# Conversation memory

"Conversation Memory" is like your saved text messages with a friend. Chat Assistants are often trained to consider past messages to help generate meaningful responses. In fact it may help to glance at the [vicuna dataset](https://huggingface.co/datasets/junelee/wizard_vicuna_70k) to understand what a model sees during training.

## Diagram

![History](https://d2uusema5elisf.cloudfront.net/courses/langchainjs-bootcamp/module_03/lesson_03.02/public/assets/history.png "History")

:::tip
You can easily verify conversational memory by introducing yourself with your name and then asking the model what your name is.
:::

The good news for us is that the ai sdk already comes with a history solution on the clientside. The entire history is sent to our defined endpoint in `messages`.

This means we can easily access `messages` from the Request.

```js { actualFilename=./protected/src/routeMemory.js endLine=35 startLine=35 statedFilename=src/app/api/chat/route.js url=https&#x3A;//gitlab.com/fullstackio/books/langchainjs-bootcamp/module_03/lesson_03.02/protected/src/routeMemory.js#L35 }
  const { messages } = await req.json()
```

Now we can take the last value as the `currentMessage`.

```js { actualFilename=./protected/src/routeMemory.js endLine=36 startLine=36 statedFilename=src/app/api/chat/route.js url=https&#x3A;//gitlab.com/fullstackio/books/langchainjs-bootcamp/module_03/lesson_03.02/protected/src/routeMemory.js#L36 }
  const currentMessageContent = messages[messages.length - 1].content;
```

Hold off on the other `messages` for now because we need to define how they interact with the prompt.

## ChatPromptTemplate

Langchain offers support for chat models. They provide a `ChatPromptTemplate` for use with typed messages and a placeholder for memory.

Go ahead and import these:

```js { actualFilename=./protected/src/routeMemory.js endLine=8 startLine=7 statedFilename=src/app/api/chat/route.js url=https&#x3A;//gitlab.com/fullstackio/books/langchainjs-bootcamp/module_03/lesson_03.02/protected/src/routeMemory.js#L7 }
import { ChatPromptTemplate, MessagesPlaceholder } from "langchain/prompts";
import { AIMessage, HumanMessage, SystemMessage } from "langchain/schema";
```

We can define our Prompt Template like this.

```js { actualFilename=./protected/src/routeMemory.js endLine=51 startLine=46 statedFilename=src/app/api/chat/route.js url=https&#x3A;//gitlab.com/fullstackio/books/langchainjs-bootcamp/module_03/lesson_03.02/protected/src/routeMemory.js#L46 }
  const prompt = ChatPromptTemplate.fromMessages([
    ["system", systemTemplate],
    new MessagesPlaceholder("chat_history"),
    ["human", humanTemplate],
    ["ai", "Let's think step by step."]
  ])
```