Beating BERT? Small LLMs vs Fine-Tuned Encoders for Classification
I ran 32 experiments comparing small LLMs to BERT on classification tasks. Turns out 2018-era BERT is still really good at what it does.
The Case Against pgvector
What happens when you try to run pgvector in production and discover all the things the blog posts conveniently forgot to mention
A Production Framework for LLM Feature Evaluation
An empirical analysis of LLM application patterns that successfully scale in production systems, focusing on extraction, generation, and classification use cases
A Computer Made This
OpenAI's 4o image generation is a step change in AI capabilities. A look at what reasoning in pixel space means for creative work.
RAG: From Context Injection to Knowledge Integration
A technical dive into the limitations of current RAG approaches, examining architectural challenges and exploring pathways to more integrated knowledge-aware LLM architectures.
Deep Dive into Python Async Programming
A deep dive into Python's asynchronous programming model, exploring the event loop, coroutines, tasks, futures, and the under-the-hood mechanisms that make it all work.
Mastering Integration Testing with FastAPI
Integration Testing FastAPI: Harnessing the Power of Mocking Backend Services with MongoMock, MockS3, and More
CheeseGPT
A (Very) Simple RAG Tutorial
Effective Error Handling
A simple an effective approach for handling end users who struggle with errors
Running Jupyter lab behind NGINX--Part 2
Part 2: Configuring Jupyter Lab authentication behind NGINX, handling token-based auth bypass, and securing the reverse proxy setup on EC2.