AI Tutorials
Practical AI development tutorials
Articles
![How to Build an AI Agent with Per-User OAuth Access [Full Handbook]](https://cdn.freecodecamp.org/platform/universal/fcc_primary.svg)
How to Build an AI Agent with Per-User OAuth Access [Full Handbook]⭐8
Learn to build an AI agent with per-user OAuth access for Slack and GitHub. Secure identity handling, token flow, and practical implementation.
Before Full Agentic RAG: Know How You Decide, and the Parsing⭐9
Before full agentic RAG, control parsing choices with an explicit dispatcher: plan, execute, and log each method for transparent, validated enterprise retrieval...
Backpropagation Explained for Beginners (Part 3): How⭐9
Learn how backpropagation really works in neural networks, explained simply with chain rule gradients, shared derivatives, and step-by-step examples for beginne...
Building Multimodal Workflows with a Local LLM⭐10
Build a multimodal LLM workflow with Gemma 4 on Ollama to analyze photos locally, transforming images into structured, searchable records for privacy-focused ap...
How to Place Vertiport Locations in Any City Using Geospatial⭐10
Use geospatial ML to place vertiports in any city, combining population data with screening rules for safer, optimized air taxi hub locations.
Product Experimentation at Scale: How Airbnb, Netflix, Lyft, and⭐7
How Airbnb, Netflix, Lyft, and Uber use causal inference to measure LLM feature impact at scale—revealing key techniques and pitfalls.
Should AI Developers Make the Switch from Polars to Pandas?⭐10
Explore the key differences between Pandas and Polars for AI development, including performance, design philosophy, and when to switch.
The Budget Split That Explains Itself⭐10
The split explained: why one optimization rule quietly inflates budget cost — and the modeling choice that hides it.
Can a Local LLM Run My AI Assistant? A 27-Task Replay Test⭐10
Can a local LLM power an AI assistant? A 27-task replay test found a 3×RTX 3090 setup with a 122B model scored 80/100—787× cheaper than Claude.
Product Experimentation with Doubly Robust Estimation: When Both⭐8
When both your propensity and outcome models are wrong, doubly robust estimation keeps LLM experiment results consistent—if at least one model holds.