Skip to content
Tarsoft

AI & Automation

LangChain: Build Smart AI Apps Powered by Your Own Data

Learn LangChain's building blocks — prompts, chains, memory, RAG and agents — and build a chatbot that answers from your own documents.

  • Beginner to Intermediate
  • 2 days
  • HRD Corp claimable

Overview

Large Language Models like GPT, Claude and Gemini are changing how applications are built. LangChain is one of the most widely used frameworks for connecting these models to prompts, data, tools and memory.

In this 2-day hands-on training, participants learn LangChain’s core building blocks, from prompt templates and chains to Retrieval-Augmented Generation (RAG) and agents. Every session has a lab, and the training ends with participants building their own RAG chatbot that answers questions from their own documents — a strong foundation before moving on to LangGraph and advanced agents.

Delivery is physical (classroom) or live online: about 30% concepts and demos, 70% hands-on labs, with code notebooks provided.

The course is for software developers starting with AI application development, data analysts and data scientists who want to build LLM-powered tools, IT professionals and tech leads exploring generative AI for their organisation, and fresh graduates with basic programming knowledge. You need basic Python; no AI or machine learning background is required. If you do not have an LLM API key, Tarsoft can provide a shared training key.

Participants leave with the source code and notebooks for every lab, a working RAG chatbot, a prompt template and RAG starter kit, and a Certificate of Completion. Join a public session, or run it in-house for your team, on-site or online. It is HRD Corp claimable, subject to grant approval; see how HRD Corp claims work.

After the course

LangGraph picks up where this course ends, with multi-step and multi-agent workflows. To run AI on your own machines instead of a cloud API, see Python Intermediate with Ollama. For team-wide habits around AI tools, see Prompt Engineering.

Schedule is tentative and subject to change based on the trainer’s delivery, participants’ progress, and training needs.

What you’ll learn

  • Build LLM-powered applications using LangChain and Python
  • Design effective prompts and reusable prompt templates
  • Create multi-step chains using LCEL
  • Extract structured data from unstructured text
  • Build chatbots with conversation memory
  • Develop RAG systems that answer questions from their own documents
  • Create simple agents that use tools to complete tasks

Who should join

  • Software developers and engineers starting with AI application development
  • Data analysts and data scientists who want to build LLM-powered tools
  • IT professionals and tech leads exploring generative AI for their organisation
  • Fresh graduates and students with basic programming knowledge

Prerequisites: Basic Python (variables, functions, lists and dictionaries, installing packages with pip). No prior AI or machine learning experience required. Bring a laptop with 8GB RAM, Python 3.10+, VS Code or Jupyter, and an LLM API key (OpenAI, Anthropic or Google Gemini) — Tarsoft can provide a shared training key if needed.

Course modules

  1. 01   LangChain Fundamentals

    What LangChain is and when to use it, the ecosystem (LangChain, LangGraph, LangSmith), common use cases, and setting up a working project.

  2. 02   LLMs & Chat Models

    LLMs vs chat models, message types (system, human, AI), temperature and max tokens, switching providers with minimal code change, and streaming responses. Lab: your first LLM call.

  3. 03   Prompt Templates

    Prompt engineering basics, PromptTemplate and ChatPromptTemplate, variables, few-shot prompting, and reusable prompts.

  4. 04   Chains with LCEL

    The pipe syntax, Runnables, sequential and parallel chains, and passing data between steps. Lab: build a summarise-and-translate chain.

  5. 05   Output Parsers & Structured Output

    String and JSON parsers, structured output with Pydantic, and handling parsing errors. Lab: build an information extraction app such as a resume or invoice extractor.

  6. 06   Memory & Conversation History

    Why LLMs are stateless, chat message history, trimming and summarising long conversations, and session-based memory. Lab: build a chatbot with memory.

  7. 07   Document Loaders & Text Splitting

    Loading PDFs, Word files, web pages and CSVs, metadata, chunking strategies, and choosing chunk size and overlap for different documents.

  8. 08   Embeddings & Vector Stores

    What embeddings are, semantic similarity, embedding models, vector stores (Chroma, FAISS), and similarity search. Lab: build a semantic search.

  9. 09   Retrieval-Augmented Generation (RAG)

    The RAG architecture, retrievers, building a RAG chain, citing sources, reducing hallucination, and basic RAG evaluation. Lab: ask questions to your PDF.

  10. 10   Tools & Agents

    Creating custom tools, tool calling, how agents choose a tool, when to use a chain, RAG or an agent, and an introduction to LangGraph. Lab: build an agent with custom tools.

  11. 11   Hands-on Project: RAG Chatbot

    Combine loaders, embeddings, retrieval, prompts and memory into an end-to-end RAG chatbot, then demo it answering questions from real documents.

FAQ

+ Is this course HRD Corp claimable?

Yes, subject to your grant approval. Tarsoft is a registered training provider and will help with the application.

+ Do I need my own AI API key?

It helps, but it isn't required — Tarsoft can provide a shared training key. A setup guide is sent before the training.

+ What do I take home?

Source code and notebooks for every lab, a working RAG chatbot project, slides and reference materials, a prompt template and RAG starter kit, and a certificate of completion.

+ What comes next?

The LangGraph course picks up where this course ends, moving on to multi-step and multi-agent workflows.

Related courses

← All courses

WhatsApp

What can we help with?

Answer two quick questions so the right person replies with what you need.

Topic

We reply Mon–Fri 9am – 6pm and Sat 9am – 12pm.