Skip to content
Forward Deployed
Forward Deployed
LLM Application Frameworks: LangChain & LangGraph: sections and modules
T4.1

LangChain Core: Models, Prompts & Parsers

Call commercial and open-source models natively and through LangChain.

Lessons
9 lessons
By
Sourav Banerjee
Prerequisites
None

Lessons

Start module
9 lessons
  1. 1
    LangChain Core: Models, Prompts & Parsers: overview

  2. 2
    LangChain introduction and the 1.x package split

  3. 3
    Commercial LLMs natively and with LangChain

  4. 4
    Open-source LLMs: Transformers, Inference Client, Groq, LangChain

  5. 5
    Inputs and outputs: messages, LLM vs chat model

  6. 6
    Prompt templates and few-shot composition

  7. 7
    Output parsers

  8. 8
    LangChain 1.x recap course, ending in a structured-output Q&A bot

  9. 9
    LangChain Core: Models, Prompts & Parsers: interview questions

Practise and Check

Deliverable
Structured answers: build a prompt template and parser that turn a question plus three passages into a JSON answer with a citations list.
Assessment
Answer the five Interview questions without notes, then run the lab.
Interview probe
Larkspur's refund answers start failing: about one request in fifty raises `OutputParserException` after a model upgrade. The previous model never did. Where do you look first?
Common mistakes
  • Migrating ConversationBufferMemory by changing only its import path
  • Shipping an init_chat_model provider swap without checking the golden questions
  • Typing citations as list[str], so invented passage ids pass

Your Progress

How confident are you with this module?

1 means "I could not explain it yet"; 5 means "I could teach it". Rate again any time.

 

Saved in this browser when you leave the box.

Related Modules