This course will teach you to design, build, and optimize AI chatbots using transformers and RAG through hands-on projects and Streamlit UI.

Building Chatbots with Open Source LLMs_ Langchain and Streamlit.png

Dockerfile-NLP.tar.gz

GROQ_API_KEY

Docker-NLP

Docker-NLP-Debug

text_classification

table_question_answering

translation

summarization

sentence_similarity

text_generation_inference

retrieval strategies

vector-stores

langchain_chains_QA

sentiment-analysis

Streamlit

This course provides a comprehensive guide to designing, building, and optimizing chatbots using modern AI technologies. It begins with an introduction to chatbots, foundational concepts in AI, and the mechanics of transformer models, equipping you with a solid understanding of how natural language processing underpins chatbot systems.

You’ll learn to implement large language models (LLMs) for tasks such as summarization, translation, and inference alongside data preparation techniques and retrieval-augmented generation (RAG) workflows using tools like LangChain. You’ll also cover advanced concepts like prompt engineering, retrieval chains, and developing user interfaces with Streamlit.

The course concludes with an integration and evaluation phase, where you’ll deploy your chatbot systems and assess performance. A hands-on capstone project solidifies these skills, making this course ideal for developing robust, user-friendly chatbots using cutting-edge AI methodologies.

Build AI Chatbots with Open-Source LLMs, LangChain, and Streamlit

To require extensive training data for every new task

To generate multiple questions from a single prompt

To apply the model's pretrained knowledge without specific examples

To enhance the model's ability to understand long paragraphs

What is the primary purpose of zero-shot prompting?

What is the primary purpose of zero-shot prompting?

Which prompting strategy is useful for handling complex reasoning tasks by generating multiple reasoning paths?

Which prompting strategy is useful for handling complex reasoning tasks by generating multiple reasoning paths?

Directly answer user queries without additional steps

Reduce the emotional content of the responses

Integrate reasoning paths with actionable steps

What is ReAct prompting designed to achieve?

What is ReAct prompting designed to achieve?

They offer first-class streaming support and async support.

They do not allow modifications once created.

They are exclusively for handling non-language tasks.

They use legacy methods that are efficient.

What is a key feature of the LCEL Chains in LangChain?

What is a key feature of the LCEL Chains in LangChain?

It passes data through without alteration.

It executes custom functions within the chain.

It processes textual data into embeddings.

What is the function of the `RunnableLambda` class in a LangChain chain?

What is the function of the <code>RunnableLambda</code> class in a LangChain chain?

Enhanced operational efficiency through parallel execution

What is a key advantage of using LCEL chains over legacy chains?

What is a key advantage of using LCEL chains over legacy chains?

By integrating with machine learning frameworks

By allowing manual coding for each operation

By providing input and output schemas for each chain

How does LCEL support complex chain construction?

How does LCEL support complex chain construction?

Test your understanding of prompt engineering and retrieval chains.

Introduction to Building Chatbots

Understanding Transformers

Understanding Large Language Models (LLMs)

Data Collection and Preparation

Optimizing RAG Workflows with LangChain

Prompt Engineering and Retrieval Chains

Chatbot User Interface Development with Streamlit

Chatbot Integration and Evaluation

Capstone Project

Conclusion and Future Developments

Quiz: Prompts and Chains