Designers can't control every model output, but we can design the conditions that shape generation. Over the last year I've constructed a framework for myself to make sense of how AI is changing our roles, what we work on, and what we can influence.
(full post in the comments)
Boris Cherny (Claude Code's creator): by 2026 the engineer who can't throw away last month's work is dead weight
we don't build for the model of today, we build for the model six months from now
50 minutes with Boris Cherny, and almost none of it is what you'd expect.
the part that reframed how I think:
there is no piece of Claude Code that existed 6 months ago. it's been written, thrown away and rewritten over and over. nothing survives.
his actual loop:
→ build the tool for yourself first
→ give it to users, talk to them, learn
→ sometimes you land on a good idea. sometimes you don't
the line senior engineers need to hear:
"I'm wrong probably half the time. half my ideas are bad." being wrong fast used to be a founder skill. now it's the engineer skill.
the whole thing is a masterclass in building for where the model is going, not where it is.
worth every minute
for anyone asking where to learn this stuff:
• RAG → https://t.co/4bzbUIwV5g
• Agentic RAG → https://t.co/IotOiGmV1Y
• AI Agents → https://t.co/nEeMnVJQbk
• Multi-Agent Systems → https://t.co/pavDPVJEFj
• LangGraph → https://t.co/3miEqqFzF0
• LangGraph (code) → https://t.co/v7kxHZXqba
• MCP → https://t.co/lKawRb4etX
• Memory Systems → https://t.co/LSaT2UaPAS
• Evals → https://t.co/vxChxa1kqQ
• Context Engineering → search "Context Engineering Survey" on arXiv
and please skip the "build an ai agent in 10 minutes" videos
build something, watch it fail, then figure out why.
Anthropic engineer:
"You're not supposed to prompt Claude. You're supposed to build a system that prompts itself."
this is one of the best workflows I've seen in a long time
in this video she breaks down exactly how most people are using Claude:
- the daily workflows Anthropic's own engineers automated first
- the task pipelines most users don't know Cowork can run
- the scheduling system that handles your busywork while you do real work
- why opening Claude to type one prompt at a time is the 2024 way of doing things
if you've been using Claude for more than a month and never left the chat window, you've been using one agent when you could be running a team of them
instead of another show tonight, watch this
make sure to bookmark it before it gets lost in your feed
the guide is in the article below
Claude Code creator:
"I don't prompt Claude anymore. I write loops - and the loops do the work. My job is to write loops."
in 30 minutes Boris reveals his actual daily Claude Code setup.
Claude Code + loops + dynamic workflow
Worth more than a $500 vibe-coding course
The new "Vibe Search" mode is one of the first beta features shipped behind the Research Lab toggle.
Think of it as Same Vibe, but search driven. It has quickly become one of my favorites.
We're shipping the onboarding for this feature a little later, in the meantime I'll let you explore and guess for yourself how its different to the regular search (:
Typography lovers will love this.
→ https://t.co/8IlNV1TA5u
Think Spotify, but for typography.
Discover typefaces from foundries around the world.
Also explore:
• a node-based graph of designers & foundries
• a globe of top foundries and typography activity
A beautiful way to explore type.