In partnership with
Hey there 👋 This week’s issue is all about execution. Not vague ideas or future promises. We’re exploring real tools, agents, and systems that help you ship faster, market smarter, and scale (plus, your favorite tools just got a lot stronger).
.
In this issue:
Figma unlocks a new era of AI-powered development 🧠
This agent built a full app and fixed its own bugs 🛠️
Why most new founders are playing the wrong game 📘
The 4-step system that took an app to $44K MRR 📈
Notion just hired an AI team so you don’t have to 🤖
Typing is a thing of the past
Typeless turns your raw, unfiltered voice into beautifully polished writing - in real time.
It works like magic, feels like cheating, and allows your thoughts to flow more freely than ever before.
With Typeless, you become more creative. More inspired. And more in-tune with your own ideas.
Your voice is your strength. Typeless turns it into a superpower.
News
credit: Figma
TL;DR: Figma just turned its MCP (Model Context Protocol) server into an API for AI. Remote access now lets agents and dev environments connect to live design and code context inside Figma—including Figma Make, their AI-first app builder. This bridges the gap between design artifacts and production-ready software.
The MCP server can now serve remote AI models, browser-based tools, and IDEs like VS Code with structured, queryable design and code context.
Figma Make files are now accessible via MCP, letting AI agents read real code instead of reverse-engineering visuals.
Code Connect gets a major upgrade with in-app component mapping, aligning Figma design elements with production code and boosting AI-generated output quality.
Partners like Cursor, Anthropic, Replit, Warp, and Windsurf are already live; third-party MCP servers and REST API updates are on the roadmap.
This turns Figma from a design tool into a programmable surface. This means your design system, UI logic, and code structure can now drive real-time generation, editing, and even refactoring. If you're building anything with agents, copilots, or codegen, Figma just became a full-stack context engine. The design layer is no longer a spec—it’s an interface.
AI
TL;DR: In a demo from MakerThrive, Emergent shows off a fully autonomous AI agent that builds and deploys full-stack apps from a single prompt. Unlike most “vibe coding” tools, Emergent asks clarifying questions, handles version control, backend tests, and GitHub sync without having to integrate other tools.
You prompt the agent with your app idea (including features, target users, and success criteria), and it builds the frontend, backend, auth, and database logic, automatically.
Emergent asks clarifying questions before it starts, avoiding hallucinated builds or wasted credits on bad assumptions.
The agent runs functional tests as it builds, catches and fixes errors on the fly, and even supports rollbacks and GitHub sync.
You can fork projects, run multiple builds at once, and turn the output into a mobile app for iOS or Android—no extra setup.
Emergent is a powerful tool that acts as your execution layer. Most AI dev tools are still replying on you to debug, test, and connect the pieces (ugh). Emergent closes that loop. If you’re building an MVP or apps, this means more shipping and less troubleshooting. Let us know what you end up building/cloning!
Productivity
credit: Notion
TL;DR: Notion just launched its most ambitious upgrade yet: Notion 3.0, with fully autonomous AI agents that don’t just assist, but actually do your work. Agents can execute 20-minute workflows across docs, databases, and integrations, with memory, personalization, and soon, full teams of specialized agents.
Notion Agents can now create docs, update databases, search across tools, and execute multi-step actions—all within your workspace.
Agents are fully customizable via Notion pages that act as instruction memory banks, letting you control behavior, style, and preferences.
Early use cases: synthesizing cross-platform feedback, maintaining knowledge bases, generating reports, and automating onboarding.
Custom Agents (coming soon) will let teams deploy multiple autonomous agents on schedules or triggers (so you can spend more time adding pretty icons to your notion docs and not prompting).
This is a big leap from “AI in Notion” to “AI using Notion.” Instead of giving you suggestions or summaries, these agents complete tasks—searching, compiling, writing, updating, and even collaborating. If you're drowning in admin, research, or ops work, Notion just gave you an automation layer built right into your second brain.
Master Marketing Measurement with Incrementality
Still guessing which campaigns actually work? Discover how leading brands cut waste and scale smarter with incrementality. This free guide breaks down how to measure true impact, optimize spend, and move beyond outdated attribution. Download now and start making smarter, data-backed marketing decisions.
Founder Story
If you had to start a business in 2025, here’s the smartest playbook
TL;DR: Daniel Priestley lays out a roadmap for building a business in 2025: skip the “big idea,” apprentice under real founders, and sell high-value services to people who can pay. We’ve said it before and we’ll say it again: you need a strong framework to build a lean, profitable business that actually fits how people live and work today.
Traditional education trains compliance, not initiative. Entrepreneurship requires the opposite: vision, risk tolerance, and autonomy.
Start by working inside a successful small business to learn the game before playing it yourself. Priestley calls this the “766 model.”
Real businesses start with paying customers, not products. Test demand before you build.
Lifestyle businesses (3–12 people, $100K–$2M+ profit) are not a compromise—they’re the new success model for digital-first founders.
It’s easy to overcomplicate things. I’m sure most of you have done it at least once when building. It’s easy to chase volume, try to “think different”, or burn months perfecting something that no one asked for. Priestley’s method is the complete opposite: build boring, sell premium, and optimize for freedom. This is practical, repeatable, and aligned with how solopreneurs and agencies are successful in 2025.
Tutorial/Framework
The zero-to-10K app user playbook, from someone who hit 1M
TL;DR: Steven Cravotta, who scaled his app to over 1M users and $44K MRR, breaks down the exact four-step system he used to get there. His framework combines Reddit discovery, viral content, low-risk creator marketing, and LTV-driven paid ads to help early-stage apps escape the “no users” trap.
Reddit is your launchpad—engage niche subreddits to earn early users and feedback, not just clicks.
Viral messages come from focusing on the transformation, not the tool. Study what’s working on TikTok before creating.
Use Posted to hire creators on a performance basis. You only pay when content performs, and you own the rights.
Scale winning content with paid ads and track LTV so you know exactly how much you can spend to grow.
Cravotta’s system gives you a replicable path: validate demand in niche communities, test viral hooks fast with creator content, and double down only on what works. This lets you grow without burning budget or guessing blindly. If you're stuck at launch or stalled at 500 users, this framework shows exactly what to do next.
Want to sponsor this newsletter? This newsletter goes out to thousands of creators, builders, and early-stage founders every week. If you’d like to get your product or brand in front of them, you can book a sponsorship slot here
