These concepts are critical for responsible AI product development—understanding failure modes, safety measures, and evaluation approaches directly informs design decisions. Safety Concepts AI Safety The interdisciplinary field preventing accidents, misuse, or harmful AI consequences. Encompasses alignment research, risk monitoring, robustness, and developing norms for beneficial AI operation. Reference: Amodei,
Understanding how models are built and adapted helps designers recognize capability sources, appreciate customization options, and understand training-related behaviors and limitations. Pre-Training Concepts Pre-training The initial, computationally intensive phase of training on massive unlabeled text. During pre-training, models learn language patterns, grammar, factual knowledge, and reasoning capabilities by processing billions
Understanding how LLMs are built helps designers grasp capability limits, performance tradeoffs, and why certain behaviors emerge. This section covers the core concepts that explain how modern AI models work. Core Architecture Transformer Architecture The neural network design that revolutionized AI, introduced in 2017's "Attention Is All
Understanding the AI landscape helps designers navigate vendor decisions, cost considerations, and market dynamics. Here are a bunch of foundational concepts that we should all have some awareness of. No need to learn them by heart, but this is a good reference manual with references where possible. Foundation Models Large-scale
The agentic paradigm represents a fundamental shift from reactive chatbots to proactive AI systems that can plan and execute multi-step tasks. AI Agents / Agentic AI Systems autonomously planning, deciding, and acting to accomplish goals—moving beyond single-response chatbots to multi-step problem solvers. Combine LLMs (reasoning), tools (actions), and memory (context)
A few months of building Agent Skills in practice. Here are the small things I wish I'd known earlier - from packaging correctly to using skills as a lightweight knowledge base.
Free software that runs open source AI models locally on your computer. Unlimited access to models from OpenAI, Google, Meta, and others with no usage limits, no fees, and complete privacy.
There's a feature quietly baked into the Claude Chrome extension that hints at something big for teams building AI-assisted workflows. Here's what it is, and why designers should pay attention.
Agent Skills became an industry standard almost overnight. Here's what designers need to know about this new way to teach AI repeatable workflows, and why it matters for design practice.
Talk to Claude about your reading list. Organize hundreds of bookmarks in seconds. Built in a weekend by a designer who doesn't write code.
Upload a screenshot, get a link, paste it into Slack. No upload wait, no file size limits, no attachments buried in thread history. CleanShot X's cloud sharing keeps everything fast and accessible.
Using SUS to evaluate a retail POS system taught me that the right metric at the right time gives you leverage, especially when users have a shared baseline for comparison.