AI
Artificial intelligence explained simply: how it works, where it helps and how to use it well. Showing page 1 of 2 (15 total articles).
Do you really need a large language model? When a smaller solution wins
I often get asked if an LLM is mandatory. The answer depends on the task, data, and constraints—here’s how to decide.
What actually runs when you load a model on Hugging Face
A look at the hidden execution path when pulling a model from Hugging Face and how to keep your environment safe.
Why 8-bit quantization can halve LLM size and how outlier weights spoil it
8-bit quantization can shrink a language model dramatically, but a single outlier weight can break everything. Here’s how to do it right.
How to Benchmark AI Agents That Use Tools and Follow Instructions
A hands‑on guide to building a reliable benchmark for AI agents that must invoke external tools and respect safety constraints.
Why Large Language Models Forget Facts and How to Keep Them Honest
LLMs often hallucinate. I explain why and give concrete ways to keep them honest.
How to Keep Your AI Agent From Doing Something Terrible
I once watched an AI assistant delete a production database; here’s how to avoid that nightmare.
Why AI audit logs are unreliable and how to improve their trustworthiness
AI audit logs look solid on paper but hide many traps. I explain the real problems and give actionable ways to make them reliable.
Measuring Speech-Recognition Bias with Whisper on African Accents
A step‑by‑step guide to testing Whisper on African speech, spotting errors, and making the results reproducible.
Understanding Recursive Self-Improvement in AI and Its Practical Limits
Recursive self‑improvement is the idea that an AI can upgrade itself. I break down the concept, the mechanisms, and the practical hurdles you’ll hit.
Need help building a website?
Send me a message and let's talk.