Up-to-date web scraping combined with LLMs.
12 lessons
AI engineer content. Only what you need (not more, not less). New videos every week!
This lesson demonstrates building a conversational AI agent using Supabase edge functions and Langchain, integrating OpenAI and external APIs for enhanced functionality. The agent manages conversation state, interacts with a CLI, and leverages a Supabase database for persistent memory, showcasing a practical serverless application.
This lesson demonstrates building cost-effective conversational agents using Supabase Edge Functions, emphasizing the crucial role of persistent storage and thread IDs for maintaining context across multiple interactions. By leveraging Supabase's database and Edge Functions, developers can create efficient and scalable agent workflows that retain conversation history and metadata, ensuring smooth and continuous user experiences.
This lesson demonstrates building an AI-powered sales workflow using OpenAI's Agents SDK. It covers multi-agent collaboration, data extraction from LinkedIn, personalized email generation, and efficient task delegation through agent handoffs.
Master the OpenAI Agents SDK to build sophisticated AI agents that seamlessly integrate multiple tools and maintain context across interactions. Learn how to leverage features like lifecycle hooks, guardrails, and easy tool integration for efficient and powerful agent development.
This lesson teaches you to build robust AI agents using LangChain's LangGraph Functional API, simplifying complex workflows with features like human-in-the-loop feedback and checkpointing for seamless interruption and resumption. Master agent-based system design, leveraging functional programming principles and LangGraph's powerful workflow management capabilities for creating adaptable and resilient AI applications.
You need to trace and eval, otherwise you're not in production.
4 lessons
This lesson teaches you how to use Firecrawl for web scraping, focusing on extracting event data from websites like Meetup, Amazon, and Perplexity. It also demonstrates how to combine web scraping with LLMs for OCR and data analysis, overcoming challenges like unreliable website structures and improving data extraction accuracy.
This lesson explores AI-powered web scraping tools, comparing user-friendly platforms like ScrapeGraphAI and Tavily with the more advanced, customizable Firecrawl API. Key features, pricing models, and integrations with LLMs are discussed to help viewers choose the best tool for their data extraction needs.
This lesson showcases Firecrawl, a user-friendly web scraping SaaS, integrating seamlessly with Python and LLMs for efficient data extraction. Learn how to quickly onboard customers by automating data retrieval from websites like Whit's Custard, leveraging Firecrawl's frequent updates and ease of use for streamlined workflows.
This lesson teaches you how to use Python's Firecrawl package to scrape pricing data from Stripe and Paddle, then uses a large language model to compare their features and costs. The tutorial covers web scraping techniques, LLM integration, and a detailed analysis of Stripe and Paddle's pricing models, highlighting key differences and advantages.
This lesson teaches efficient web scraping using Firecrawl and webhooks, eliminating inefficient polling methods. It demonstrates building a FastAPI server to receive real-time job completion notifications from Firecrawl, enabling asynchronous processing and analysis of scraped data, such as comparing Stripe and Paddle pricing with an LLM.
This lesson teaches you how to rapidly evaluate software vendors using Python, leveraging the Firecrawl API to automate web scraping and either Groq or OpenAI to analyze gathered data against predefined criteria (like SOC 2 compliance and F500 testimonials). The process dramatically accelerates the software selection process compared to manual research, focusing on platforms like Drata, Vanta, and Secureframe.
This lesson explores the burgeoning trend of web scraping in 2024, driven by the need for real-time data to enhance LLMs and search engines. It demonstrates various web scraping techniques, including using Python libraries and AI-powered tools, while analyzing their cost-effectiveness and comparing the resulting data quality for optimal LLM performance.
This lesson teaches you to build efficient web scraping agents using Python libraries `langgraph` and `firecrawl-py`. By combining agent-based design with a state machine, you'll learn to extract specific product information from websites like Canada Goose, handling errors and optimizing for speed.