Fine-tuning — training an existing foundation model on a custom dataset to change its behaviour — is often the first thing teams reach for when a gene
2026-04-27
Building an AI application with Claude that works reliably in production requires more than calling the API and displaying the response. Here are the
2026-04-25
Agentic AI — AI systems that take sequences of actions to complete goals, rather than just responding to single prompts — has matured significantly in
2026-04-20
Building an LLM-powered application is the easy part. Knowing whether it is working well — and detecting when it regresses — is the hard part. LLM eva
2026-04-18
Open science — open access, open data, open methods, preregistration — has moved from fringe advocacy to mainstream expectation in most fields over th
2026-04-18
Multimodal AI — models that process both text and images (and increasingly audio and video) — has moved from research novelty to production capability
2026-04-17
Prompt engineering — the practice of writing better inputs to get better outputs from AI models — has evolved significantly since the GPT-3 era. Here
2026-04-16
AI coding tools have progressed from basic autocomplete (GitHub Copilot, 2021) to agents that can take a task description, write code, run tests, and
2026-04-14
When building an AI application with access to domain-specific knowledge, there are three main approaches: prompting (include the information in the c
2026-04-13
AI agents are the most consequential development in AI since the large language model breakthrough. Here is what they actually are, what makes them di
2026-04-11
