thenewboston
LLM Application Development and Python Web Stacks with a focus on practical API implementation.
Nutrition Label
This channel provides highly accessible, step-by-step coding tutorials that prioritize practical implementation over abstract theory. The content excels at demonstrating the 'happy path' for integrating APIs like OpenAI and LangChain, making it ideal for beginners looking to get their first script running. However, the analysis rarely deviates from standard documentation or explores complex edge cases.
Strengths
- +
- +
- +
Notes
- !Tutorials focus on basic implementation and may not cover complex production errors.
- !Code examples are functional but strictly follow standard documentation patterns.
Rating Breakdown
Breakdown across the key dimensions we rate. Methodology →
Recent Videos
Bonsai - 3 - Planting the Seeds
Bonsai - 2 - The Seed
Bonsai - 1 - Camera Test
The Singularity - Tutorial 1 - Architecture
Prompt Engineering for Beginners - Tutorial 19 - Building Your Own ChatGPT (part 2)
Prompt Engineering for Beginners - Tutorial 18 - Building Your Own ChatGPT (part 1)
LLM Application Development - Tutorial 5 - Long Term Memory
LLM Application Development - Tutorial 4 - Routing
LLM Application Development - Tutorial 3 - RAG
LLM Application Development - Tutorial 2 - Evaluations
LLM Application Development - Tutorial 1 - Introduction
Prompt Engineering for Beginners - Tutorial 17 - Anthropic Claude 3
Prompt Engineering for Beginners - Tutorial 16 - Anthropic
Prompt Engineering for Beginners - Tutorial 15 - Evaluation Pipeline
Prompt Engineering for Beginners - Tutorial 14 - Creating Prompt Templates
Why this rating
Evidence receipts showing why each dimension is rated the way it is.
“In this video I just want to give you guys a brief overview of how this system works... and in the next video we're going to be getting our hands dirty.”[0:18] →
Perfectly sets the scope as a conceptual primer, aligning exactly with the 'Introduction' title.
“We take that question, we look up relevant information... and then we pass both of those to the LLM.”[2:15] →
Uses a clear visual diagram to distill the complex RAG pipeline into an accessible three-step process.
“I'm going to go ahead and upload... a messy room. And for the prompt, I'm just going to say 'Describe this image.'”[04:35] →
Demonstrates the multimodal capabilities live by uploading a specific image and running the model in real-time.
“We need to pass the learning rate and the initial weights array into the seed function, otherwise it defaults to zero and the network won't learn.”[05:15] →
“Haiku is the fastest... Sonnet is the one that we are on right now... and Opus is the most intelligent model.”[00:44] →
Accurately summarizes the model family differences, but relies on standard manufacturer definitions rather than independent benchmarking.
“I'm just going to say 'tell me a joke'.”[05:45] →
The tutorial relies on a trivial 'Hello World' style example ('tell me a joke') without testing edge cases, complex reasoning, or failure modes.