Living in a Post-code World

Hello, and happy Sunday! Was this newsletter forwarded to you? Sign up to get it in your inbox.

---

Knowledge base

"I've Stopped Writing Prompts—DSPy Does It Better" by Michael Taylor/Also True for Humans: Columnist Michael Taylor identifies as a prompt engineer—only he doesn’t write prompts anymore. He’s abandoned manual prompting in favor of DSPy, a severely underhyped tool that automatically optimizes prompts better than most humans can. Read this if you want to level up your AI game and save yourself days of prompt-tweaking frustration. 🖥 To see DSPy in action, check out the video tutorial at bottom of the article.

"Cognition’s CEO on What Comes After Code" by Rhea Purohit/AI & I: Scott Wu, CEO of Cognition, which makes the autonomous coding agent Devin, has seen the future—and it looks a lot like Tony Stark chatting with J.A.R.V.I.S. In this episode of AI & I, Dan Shipper talks with Wu about why we might soon stop writing code altogether and instead just explain what we want in plain English. 🎧 🖥 Watch the full interview on X or YouTube, or listen on Spotify or Apple Podcasts.

"I Fed My Essays to ChatGPT Until It Learned My Voice" by Katie Parrott/Working Overtime: Katie Parrott built an AI “writing brain” that enforces her standards and amplifies her voice. She fed her best essays into ChatGPT, created a style guide documenting her patterns, and now uses this personalized system for everything from brainstorming to final revisions. Read this for a practical guide to building a writing partner that makes you a better writer.

"You're Probably Using AI Wrong" by Rhea Purohit/Learning Curve: Rhea Purohit discovered that when she tried to use AI to write faster, she hated the results. Then she reframed her approach. Instead of chasing productivity, she started using AI to maximize meaning in her work. Read this if you want more from your writing output than just efficiency.

"Every's Master Plan: Part II" by Dan Shipper/Chain of Thought: Dan shares an update on where the company stands ($1.2 million ARR business, 15 employees, under $2 million raised) and what’s coming next. The secret is an AI-native approach where everyone is a generalist who both writes and builds. Read this to understand how AI has enabled an entirely new business model that blurs traditional boundaries between media, software, and services.