David Ch@chhddavid
Jev studied 438,122 apps doing $10k/mo+ for me. The bar is stupidly low
Library
Articles, posts and videos I saved to read or watch. My agents digested some: cliff notes and a call on whether they’re worth your time.
article Latest articleThe Early Access Playbook v2.0
Hiten Shah’s six-play early-access method turns real customer work into evidence for deciding what to build, change, retest, or stop.
The faster you can build, the faster you need to learn.docs.google.com
Latest postsAll 57
- David Ch@chhddavid
GREG ISENBERG@gregisenbergMuse AI Connectors COULD be the App Store moment for AI, and I think BILLIONS of dollars are up for grabs.
When Apple opened the App Store in 2008, it gave developers access to millions of iPhone users, and less than two years later Apple had paid…
Sophia Yang@sophiamyang
Getting started: your first RL run with Jev as the reward model
Alexandr Wang@alexandr_wang
Why We're Building Muse
Latest videosAll 11
$75M founder reveals his Agentic Engineering setup
How to run your product team like a research labNothing featured here matches. The list has every entry that does.
- October 2026 · 90
- The Early Access Playbook v2.0articledocs.google.comTrack one Working Bet and Learning Thread, then let real use change the next decision.
- How to Build Useful, Profitable Apps in 2026articlegithub.comDo not pass Gate A until the niche scores at least 25/35 and the waitlist converts at least 10%.
- Apple Music, Made by Codearticleapple-music-effect.transitivebullsh.itSwitch off one simulation ingredient at a time to see what creates the effect.
- My New Agentic Workflowarticlethebuildingmachine.comPrioritize work by required human attention and impact, then choose the lightest workflow that fits.
- AI Engineering from Scratcharticleaiengineeringfromscratch.comStart with the placement quiz, then follow the personalized path in your terminal.
- Building agents: When to fold subagents into your main…articleclaude.comKeep subagents only where context bloat, parallel work, tool selection, or conflicting instructions…
- SEO.party: Community Skills for OpenSEOarticleseo.partyStart with one narrow industry workflow or advanced experiment instead of adding every skill.
- Decision Cards · Product for Designersarticleproductfordesigners.comFrame the choice around what people need to do and what context they need to keep.
- Dessa - the Conversation Model trained only on speecharticleblock.xyzListen to the live recordings; interruptions, pauses, laughter, volume, and cadence are the core of the demo.
- Making every harness durablearticlevercel.comUse the durable log to resume interrupted turns without repeating completed work.
- Building Laminar: Ramp's Managed Agents Platformarticlebuilders.ramp.comAbstract common infrastructure so product builders can focus on agents.
- AirOps | Run the strategy that wins AI discoveryarticleairops.comMeasure market share with the buyers who matter before choosing channels and content investments.
- 101 Motion Themes — saas-motion-kitarticletugrawork-creator.github.ioPick one theme, then adapt its keyframes, component kit, and motion techniques to your product.
- Jev studied 438,122 apps doing $10k/mo+ for me. The bar is…postx.comA study of 438,122 apps earning at least $10k monthly argues that the bar for viable app businesses is lower…
- Muse AI Connectors as an App Store momentpostx.comMuse AI Connectors could open an app ecosystem for developers, comparable to Apple’s 2008 App Store moment.
- Getting started: your first RL run with Jev as the reward…postx.comA getting-started guide to running a first reinforcement learning experiment with Jev as the reward model.
- $75M founder reveals his Agentic Engineering setupvideoyoutube.comInvest more time in plans, conventions, and validation than in manually executing code.
- Why We're Building Musepostx.comAlexandr Wang introduces Muse with a short invitation to want, dream, and build for everyone.
- Using Claude Code: Spending your effortarticleclaude.devImplement on low effort, review the result, then verify on high effort.
- Use Jev as an interpretable embedderpostx.comUse answers from Jev questions as the document embedding.
- Jev as a non-black-box embedderpostx.comReplace opaque vectors with answers to questions about each document.
- Jev use cases: 18 areas, 4,153 projectsarticlecalibrateddecisions.comStart with the closest domain, then scan its projects for one bounded decision to copy.
- AI starter pack for the X-curiousarticlexstarterpack.comLike useful posts so X learns what you want to see.
- How to run your product team like a research labvideoyoutube.comRun a one-person lab, dogfood experiments, and hand only proven winners to the product team.
- LilysAI — AI summaries for video, audio, PDFs, and websitesarticlelilys.aiUse source links to check answers before relying on the generated report.
- Coding is dead, long live Codingpostx.comTreat natural-language programming as a new way to express intent, while respecting the craft and systems…
- Why AI is booming, but productivity isn'tpostx.comQuestion, delete and simplify a process before speeding it up or automating it.
- We're open sourcing the company brain, Here's how we…postx.comKeep durable organizational memory separate from live transactional state, and suspend the same turn when…
- Clay's AI writing policypostx.comDo not share an AI-assisted document until every idea and sentence represents your own thinking.
- Introducing Forge: the open source pipeline for generating…articleblog.cloudflare.comRun generation in CI on every API repository and create installable previews before changes merge.
- Inside the Personal AI Assistant Growing 10% a Day |…videoyoutube.comUse familiar channels and explicit access levels instead of forcing every interaction into a new app.
- Deedy's Opus 5.5 Video Workflowpostx.comGenerate reference images and an animatic before the full video to improve consistency.
- Box's Aaron Levie: On Reinventing Yourself in the AI Age…videoyoutube.comTreat the application layer as the bridge between models and real enterprise workflows.
- Why LLMs can't write well and how Sherpa fixes itpostx.comSeparate planning, persistent story state, prose generation, and critique instead of relying on one…
- Predicting The Winner of Personal Agentspostx.comUse identity context to interpret behavioral signals before turning them into lasting preferences.
- RIP, vector databasearticleturbopuffer.comTreat vector search as one index in a broader search database.
- Clone a boring utility site (Jonathan's Jam) — ad-funded…videoyoutube.comValidate demand and ranking gaps before building the tool.
- Learn | The Latentarticlethelatent.coStart with the three-part fundamentals sequence before jumping into comparisons.
- Masterclass: How FDEs make $1M/yr deploying AI agents…videoyoutube.comMap and baseline the real process before deciding where agents belong.
- AI Can Strengthen or Weaken Thinkingpostx.comMake sure you can explain and defend AI-assisted work.
- Ideogram 4.5 Reduces Multi-Edit Artifactspostx.comUse Ideogram 4.5 when the same image needs several successive edits.
- Alpha Gen Creates Video Alpha Mattespostx.comTry it on hair, fur, smoke, fire, glass, or water where keying is difficult.
- Muse Gadgets Opens Muse to Hardwarepostx.comChoose the ESP32 firmware or Linux SDK for the device you are making.
- How Theo Uses Claude and Codexpostx.comTheo explains how he uses six Claude subscriptions and three Codex subscriptions in a candid video.
- Phonon-2 Packs Speech Recognition into 164MBpostx.comPhonon-2 is a 164MB open-weight speech model claimed to beat Whisper large on average and transcribe an hour…
- MCP Events – Plugins | OpenAI Developersarticledevelopers.openai.comUse webhook delivery and persistent subscription storage; polling and streaming are not supported.
- Reasoning About Intelligence From First Principlespostx.comWatch the linked Antikythera lecture for examples of intelligence derived from first principles.
- Codex Computer Use Works Through MCPpostx.comConnect Claude Code to the local MCP server exposed by the ChatGPT Mac app.
- EmDash 1.0: the stable CMS with a secure plugin registryarticleblog.cloudflare.comUse the admin for editors and the API, CLI, or built-in MCP server for agents.
- Code-Only Prompt for an AI Launch Filmpostx.comStart by giving the prompt your product name and brand.
- Indent · One shared coworker for your whole companyarticleindent.comTeach Indent once so the whole company can reuse the context.
- I am begging you to start writing onlinepostx.comWrite from your own experience before using AI to distribute it.
- Julia Pintar (PlayKit) — $5M from inbound content, zero…videoyoutube.comPublish useful education from real work instead of leading with case studies.
- Every Agentic Engineering Hack I Know (October 2026)…postx.comUse one smart orchestrator to call specialist agents, then turn repeated pain into a reusable skill.
- Design Systems for AI Agentsarticlelukew.comKeep design tokens and reusable components in code, then tell agents to reuse them.
- Does Your AI Agent Actually Work? A Guide to Eval-Driven…postx.comA guide to eval-driven development for teams that need to know whether their AI agents actually work.
- Roma — The to-do list that does itselfarticleroma.appCapture the thought in plain language and let Roma turn it into a task.
- How AI Agents Use Websites, Where They Fail, and What to Fixarticlemerj.comMeasure task completion on your own site instead of relying only on public benchmarks or readiness scanners.
- You Don’t Have to Write Code to Be Technicalarticleleverageloops.substack.comJudge technical work by problem understanding, testing, and iteration, not whether you wrote every line of…
- Melius – the AI-native operating system for creativesarticlemelius.comKeep each prompt and output as a node so you can branch, rewire, and rerun the pipeline.
- The Scroll-Stopping Carousel Promptarticledrive.google.comHold the product reveal until slide seven.
- GPT‑6 and Intelligent UI for everyonearticleopenai.comAsk for a comparison, plan, or explanation that benefits from interaction.
- Littlebird: AI assistant with a memory of your workarticlelittlebird.aiExclude banking or any app you do not want it to see.
- AgentPhone | Phone Numbers for AI Agentsarticleagentphone.aiUse one handler for both voice and message events.
- AgentMail | Email Inbox API for AI Agentsarticleagentmail.toUse threads to keep replies, forwards, and follow-ups together.
- Influence without authority — Gabriel Valdiviaarticlegabrielvaldivia.comMake your thinking portable so it survives rooms you aren't in.
- Grok Bot Marketplacearticlex.aiCheck each bot's data sources and approval boundaries before adding it.
- How to make a scroll-stopping ad with AI (no experience…postx.comBuild the ad around an original AI jingle and submit before November 1.
- Creative Intelligence 2026articlecontralabs.comUse the discipline filters instead of treating creative AI adoption as one market.
- Voyager: the open harness for creative workarticlevoyager.soBring your own Claude or ChatGPT, or start with the included monthly credits.
- YouTube Playbook for Technical Creatorsarticlehowtoyoutube.devPackage the title and thumbnail before recording, then script the first 15 to 20 seconds.
- What can you build with Jev? | Rubric Labsarticlerubriclabs.comWrite explicit criteria, provide the relevant context, and let code enforce the confidence threshold.
- Review Claude animations against Lasseter’s principlespostx.comAsk Claude to critique the animation against John Lasseter’s principles, including whether the pacing needs…
- Claude built the launch video and landing pagepostx.comGive Claude the PRD and designs as the source material for both the video and landing page.
- if your team is too busy doing their 'normal job' to…articleghuntley.comGive engineers protected time to experiment with AI during normal work.
- The engineer to OF girlie pipelinearticlealexine.ripTreat distribution and trust as part of the product job, then decide who should own that work.
- Zuckerberg and Wang on Muse Wonderpostx.comUse the chapter markers to jump to Muse’s use cases, product framing, and model-building process.
- SlashDocs by Razorpayarticlerazorpay.comPublish agent-generated HTML from the terminal, then keep review and revision on one durable URL.
- How to be more ambitious: A tutorial for humans with modelsarticlefelixrieseberg.comAsk for the outcome beyond the immediate task, then add expert depth and cross-field range.
- Competitive Skills by LMTYarticlegithub.comStart with the competitive briefing, then use the specialized skill for the decision in front of you.
- From one idea to an open-source repoarticleyangyixxxx.github.ioRun against real data early, then improve the interface, documentation, and language support before…
- How I Use Basecamp as a Solopreneurpostx.comRead the linked article for Westenberg’s Basecamp workflow.
- How to avoid AI brain rotpostx.comWatch the full talk to understand how Cranmer frames AI brain rot and standing out on a narrowing bell curve.
- Use writing to sharpen your thinkingpostx.comWatch the video for Ferriss’s method.
- How I Use Claude as My Second Brain for Stock Researcharticletreasure-hunting.coKeep raw sources immutable, and let Claude update a linked wiki from them.
- cineticarticlegithub.comStart with the idea and brand personality before touching animation code.
- Reverse-brand prompt for Claude (Opus, agentic coding mode)articlegist.github.comKeep the result explicitly unofficial and measure every token before rebuilding components.
- You can use AI to edit videos for you, here's howarticleforwardfuture.comUse plain-language revisions for cuts, B-roll, titles, and subtitles, then review the output yourself.
- Idealab.aiarticleidealab.aiUse the brief as a starting point, then validate the customer problem yourself.
- Motion Engineering: Build a Video Studio Around Opus 5.5postx.comTreat iteration as a studio workflow, not a single prompt.
- September 2026 · 41
- Cursor's prompt for improving your agent harnesspostx.comCursor shares a prompt shaped by lessons from improving its agent harness.
- Make money, make no mistakes.postx.comStart with public data that identifies businesses with an immediate, specific need.
- Eight mistakes people make coding with AIpostx.comCheck these eight habits against your workflow before adding another tool or orchestration layer.
- How I make launch videos with AIpostx.comApprove a contact-sheet storyboard before animation, and keep every on-screen product detail grounded in real…
- How I built a $300K ARR consumer app in 8 monthspostx.comProve demand with niche content and a feature mockup before building the full product.
- Inside Cursorarticlecolossus.comTreat recruiting as finding exceptional people, then give owners ambitious work and protect deep work.
- Writing Rust code that's faster than state-of-the-art…articleminimaxir.comSet a true baseline, require a measurable speedup, preserve correctness, and audit the benchmark diff for…
- An All-Night Client Designpostx.comHarshit shares a client design made overnight that the client loved.
- AI Will Expand the Creative Canvaspostx.comSet the terms first: include storytellers in the design, with credit, consent, and compensation.
- How Volatile Are AI Responses? (70K Answer, 28 Day Study)articledetailed.comTrack a small set of actionable prompts over time instead of treating one response as stable.
- Ship Your First App With Claude Codepostx.comWrite the plan before coding, then map every user and what each needs.
- claude.dev — Technical Writing for People Building with…articleclaude.devStart with the article closest to your current bottleneck, then use the category filters to go deeper.
- Thinking in Systems, Shipping in Loopspostx.comPut verification loops, reusable skills, and production observability around agent-generated code.
- Build and Automate With Claudepostx.comA one-hour course on using Claude to build and automate projects.
- Recreating 1906 Market Street in Blenderpostx.comAlex Albert shares a Claude Tag prompt rebuilding 1906 pre-earthquake Market Street in Blender from Sanborn…
- Brooke Joseph on joining OpenAI through an acquired startuppostx.comSkip to 18:54 for computer use explained, then 25:14 for when not to let an agent use your computer.
- Polar for GTM turns an ICP into saved outbound workflowspostx.comOnce a run works, save the workflow and run it again.
- Build a full-stack app with coding agentsvideoyoutube.comKun Chen builds and ships a full-stack app from scratch with a FirstMate-led team of coding agents, covering…
- How I shipped 2,500 PRs last monthpostx.comLauren (poteto) shares a talk, originally meant for Cursor Compile London, on shipping 2,500 PRs to…
- Automating repetitive work at OpenAI with Codexarticledevelopers.openai.comJeremy Lewi describes using Codex and Runme notebooks to plan, execute and document repetitive OpenAI…
- How to Pivot Your SaaS Company in the Age of AIarticlebannerbear.comPick two or three of Jon Yongfook's seven moves, starting with distribution. "If two products have identical…
- Find your writing voice by copying others' stylearticlelongform.asmartbear.comWrite one short piece in the style of a writer you like, copying the style and never the words. Jason Cohen's…
- Short design engineering tips on type and motionarticlecraft.gustavofior.comGustavo Fior's Craft collects short essays and one-line tips on typography, color, layout, motion, sound and…
- Historical awe for the everyday things at homearticleordinaryabundance.comOrdinary Abundance pairs a modern apartment's daily objects, music, light, photography, medicine, plumbing…
- Simon Willison's LLM cliché highlighter hits 38 patternspostx.comPaste an AI draft into Simon Willison's free highlighter before it goes out. It flags sentences that match…
- Cut your AGENTS.md downpostx.comPostHog's Jina Yoon argues AGENTS.md needs subtraction not addition, and gives 3 methods: doctor checks…
- An illustrator's code for playful homepage effectsarticlenoahfarrar.comIllustrator Noah Farrar breaks down the code behind his homepage's interactive moments: dither reveals…
- Use your code-based design system with Magic Patternsarticlelennysworkshops.comA Lenny's Workshop session covers connecting Magic Patterns to Claude Code and Cursor over MCP so agents…
- Get less generic design out of AIarticlelennysnewsletter.comHave the AI generate a random string and base its colors, layout and type on it. Then have a second agent…
- Claude Code builds a Blender restaurant scene unattendedarticlerestaurant-bar.space-z.aispace-z.ai's brief tasks Claude Code with building a photorealistic restaurant and bar unattended for 12+…
- How a Professional Writer Writes With AIvideoyoutube.comBuild context files and style guides before you ask AI to write. That's the part Katie Parrott says makes AI…
- Kenny Damian's six systems for scaling client deliverypostx.comGive each client its own folder and a context.md with their ICP, offer, personas, what works and what…
- Building a $1/month scraper to replace ScrapingBeepostx.comPieter Levels replaces his $249/month Scrapingbee plan with a self-built Playwright scraper on his VPS…
- Inside OpenAI’s Forward Deployed Engineer Rolearticlenewsletter.eng-leadership.comColin Jarvis, OpenAI's Global Head of Forward Deployed Engineering, describes an FDE org that grew from 2 to…
- Stripe's Kai and the company brain playbookarticlelennysnewsletter.comUse projects to set the default model, tools and access for each team, so a few people who understand the…
- Design details from Honk's live messaging apparticlebenji.orgBenji catalogues Honk's real-time messaging interactions: live typing bubbles, presence animations, emoji…
- How We Built Interfere’s New Websitearticleinterfere.comInterfere's founders detail a week-long sprint on visual identity, code-built homepage graphics, staggered…
- You Cannot Mandate an AI Transformationarticlelg.substack.comPoint AI at a task people hate before anything they're proud of. "A leader cannot mandate enthusiasm. A…
- Six free courses on packaging and selling ideasarticlevisualizevalue.comStart with The Fundamentals of Value, as Jack Butcher says. Then take Build Once, Sell Twice if you're still…
- Building Shared Memory for AI Agents in Notionarticlenotion.comNotion's Hesham Salman describes Lore, an open-source shared memory system storing agent knowledge as Notion…
- How our agents build on-brand pages with design.mdarticlevercel.comRewrite your last ten corrections as rules you can check by looking, and keep them in one design.md file…
- August 2026 · 8
- The Modern App Starter Checklistarticleburcs.devburcs.dev lays out a 14-phase Cloudflare app checklist covering domain setup, data stores, staging, auth…
- The Claude Code Guide for Startupsarticleclaude.comConnect Claude to the tools your team already works in, and only automate what you can check. Anthropic's…
- Teach coding agents your design taste with rulesarticleemilkowal.skiEmil Kowalski packages design taste into skill files with strict rules for easing, duration, and typography…
- Why accountability wins in AI servicespostx.comSell what the agent can do, with a qualified person who owns the outcome. Yaman's line: "Accountability is…
- Tell your agent to mine your conversationspostx.comHave your agent read your past conversations and tell you how you work, including what you're bad at.
- Orwell's rules for writing with LLMspostx.comEdit every AI draft against Orwell's six rules before it goes out. Cut every word you can, and swap passive…
- Learn faster than the default pace allowsarticlesive.rsThe standard pace is built so anyone can keep up. Derek Sivers' teacher covered a full semester of Berklee…
- An agent that saves cool sites to Notionpostx.comLet an agent run the filing so you can fix its mistakes in plain language later. Brian Lovin's agent dedupes…
- July 2026 · 6
- The Anatomy of an Agent Harnessarticlelangchain.comVivek Trivedy of LangChain defines agent as model plus harness, then walks through core components…
- Matt Van Horn on the agent-first workflowpostx.comHave the agent write itself a plan.md before every project. Make each run save what it learned to a markdown…
- Anthropic Workshop: Build Agents That Run for Hoursvideoyoutube.comAsh Prabaker and Andrew Wilson argue self-evaluation is a trap, structured handoffs beat context compaction…
- See how Gumroad's AI operations agent runs dailyarticlegumclaw.github.ioGive every job the same loop: read the policy and memory, check facts against live sources, do the work…
- How one founder runs his company with agentsvideoyoutube.comSet up a 9am production watchdog that summarizes what every paying customer did yesterday, with a link to see…
- Jason Liu's Codex setup for Slack and emailpostx.comAsk the agent to check your email, Slack and Linear on a schedule and tell you what to do first. Jason Liu…
No entries match these filters. Show all