NLP and LLM Pipelines

This module covers advanced Natural Language Processing (NLP) techniques and Large Language Model (LLM) pipelines. We’ll work with two practical applications:

  1. Structured information extraction
  2. Audio-to-blog content creation

Structured Information Extraction with LLMs

We’ll explore how to extract structured data from unstructured text using Large Language Models.

Topics covered:

  • OpenAI client setup with custom API keys
  • Prompt engineering for information extraction
  • Pydantic models for structured data
  • Schema enforcement approaches
  • Few-shot learning implementation

LLM Structured Output Notebook

Audio-to-Blog Pipeline with Local LLMs

Learn to build an end-to-end pipeline that converts audio content into blog posts using local Large Language Models.

Topics covered:

  • Podcast transcription with Whisper
  • Ollama setup for local LLM inference
  • Transcript analysis and key topic extraction
  • Blog post generation from structured info
  • Together API for advanced model usage
  • Image generation for blog illustrations

LLM - Audio to Blog Notebook

These exercises will give you hands-on experience with NLP pipelines and LLMs. You’ll learn to extract structured data, generate content, and create visuals to complement your NLP outputs.