Real work,
honest process.

Key insights & challenges

The two memory systems work completely differently and that distinction matters. Project Instructions in Claude.ai are conversational — Claude reads them silently and applies them throughout the chat. Claude Code skills are more like reference documents you explicitly invoke at the start of a session. Neither is automatic in the way you'd want: Project Instructions still drift in long sessions, and Claude Code skills still need to be called by name. The real insight is that these tools shift the burden from memory to discipline — you still have to write the rules clearly and invoke them consistently. The upfront investment is worth it, but it doesn't eliminate the need for human oversight in every session.

Key insights & challenges

The most unexpected insight came from a bias Claude introduced without flagging — it assumed gender for participants whose pronouns were never stated in the transcripts. When caught, the same assumption had been made across multiple users. For a product designed to be non-judgmental and inclusive, it was a useful reminder: AI makes demographic assumptions quietly and confidently, and human oversight isn't just a quality check — it's an inclusivity check.

Key insights & challenges

Claude lacks visual design judgment. The expand/collapse functionality it generated was structurally correct but visually wrong — the interaction felt broken and the collapsed state unpolished. I had to open Figma and draft how I actually wanted it to look. To bring the HTML into Figma I tried two tools: Magic Patterns (did not import files or add autolayout) and html.to.design plugin (converted the HTML into a proper Figma file with styles, components, and autolayout — this worked). Claude is strong at structure and logic, weak at visual feel. Figma remains the right place to make visual decisions, even when working code-first.

Key insights & challenges

The hardest part was the tooling. Git branch mismatch (master vs main), wrong Vercel framework preset causing 404s that looked like code errors, CSS module imports that work differently in Next.js, file management without terminal knowledge, image format mismatches (.jpg vs .png). Each a 2-minute fix once diagnosed — the hard part is knowing where to look. Structure first pays off immediately: tokens in one place, components in their own files is already cleaner than a single HTML approach.

Key insights & challenges

Shipping something live — even bare minimum — changes how seriously you take the next step. Prototype first, then migrate. Validate the design before investing in infrastructure.

Key insights & challenges

Two lessons from one session. First: Claude approximates visual styles from URLs — define your design system as an explicit spec before asking it to match anything. Second: Claude has no persistent memory within long sessions. The fix is Project Instructions — a persistent system prompt that eliminates re-establishing ground rules each time. Every instruction needs to be written as if it might be the only one Claude ever reads.