AI agent frameworks vs writing your own loop
Nobody has ever failed to build an agent because the loop was hard. They fail on the retries, budgets and logging a demo never needed.
Read More
Learn
Evergreen explainers, guides and comparisons. The reference material that stays true after the news cycle moves on.
Long context vs RAG: which one you actually need
A stated window is a capacity figure, not a usable-attention figure. The middle of a long input is measurably a worse place to be.
Read More
Hosted AI API vs self-hosted models: cost and control compared
The API bill is visible on an invoice. The self-hosting bill is spread across salaries, on-call rotas and slipped projects.
Read More
Postgres pgvector vs a dedicated vector database: when to switch
A second datastore is a second thing to back up, monitor, secure and be woken by. That cost is invisible in a benchmark.
Read More
Ollama vs llama.cpp vs vLLM: which to run locally
Serving one person and serving two hundred are different engineering problems. The tool that wins at one loses badly at the other.
Read More
Prompting vs RAG vs fine-tuning: how to choose
Does not know something means retrieval. Does not behave right means fine-tuning. Was not told properly means prompting, and that is most cases.
Read More
Flagship vs entry-tier AI models: when the cheap one is enough
The frontier model wins the comparison. The cheap model wins the invoice. Which matters depends entirely on your volume.
Read More
Tokenisation explained: why a model cannot count the letters in strawberry
Strawberry might be two tokens. The model is being asked to count letters inside symbols it cannot see inside.
Read More
How coding agents actually work, stage by stage
The model never touches your repository. It names a tool and supplies arguments, and your code decides whether to run it.
Read More
The self-hosted AI stack: what to run at each layer
You can run a complete AI stack on hardware you control. Whether you should depends on what an hour of maintenance is worth to you.
Read More
How multimodal AI models actually process images
The mechanism is simpler than the capability suggests, and unusually predictive of where these models fail.
Read More
Mechanistic interpretability explained: reading a model from the inside
We specify a training process, not a design. Nobody writes down what the finished network does, which means nobody starts out knowing.
Read More