We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Sarah Guo and Ryan Hoover: Dennis Crowley, "Some final words on Foursquare"
Foursquare cofounder Dennis Crowley (@dens) wrote about the company's final years after it was acquired last month, an outcome he called "disappointing." He wrote it to honor the team and to get "some closure" (essay).
Sarah Guo called it "super important honest advice to founders." Her takeaway: if you're working on a dead end, get out. Founders don't owe it to investors to keep going, because investors "regularly forgive and experience failure" and ideally care more about you doing your life's work than about a subpar outcome. She added that being responsible with investors' money is a different thing from trying to make a company you've lost faith in "work." Ryan Hoover also shared the piece ("Appreciate the retro"), pulling out Crowley's line "I was heartbroken by this ending."
Tim Ferriss: books from his new Q&A episode
Ferriss recommended several resources in his Q&A episode. Links are in the show notes.
- Ramit Sethi's podcast, which Ferriss calls "great." He borrows its question about what it would look like to invest ten times more in the one or two areas that matter most. For him those are health and relationships.
- James Clear, Atomic Habits (and BJ Fogg's work), for people stuck overthinking. Both explain how small actions add up. His advice is to start with small daily actions so you rebuild trust in yourself.
- Michael Singer, The Surrender Experiment, which Kevin Rose recommended to him as a counterexample to over-planning your life. Ferriss "very much enjoyed it" and found it "very helpful." It did leave him with questions about total versus selective surrender.
- Terry Real, Fierce Intimacy (audiobook only). Ferriss calls Real "a beacon of clarity in a world of handwavy relationship advice." If you pick one resource between this and Crucial Conversations, he says to choose this one.
- Elisabeth Kübler-Ross and David Kessler, On Grief and Grieving. It has helped friends after a death, and it is the book on grief recommended most by people he trusts. He notes he hasn't read it in ages.
He also repeated a recommendation from the last brief: Drucker's The Effective Executive, as an "old book" that is "very short" and dense with principles.
Tobi Lütke: DHH's "AI delirium" speech
At a fireside chat with Guillermo Rauch, Lütke called DHH's recent speech about "AI delirium" "a fantastic one." He gave his own reading, adding that DHH might not agree with it. In that reading, the term describes people who have moved past the p(doom) debate and are focused on bringing into the world the things that "previously were impossible." Lütke says he is at that stage or close to it.
Delian: Wajahat on Varda and "space drugs"
Delian Asparouhov said it was the "first time I've seen someone understand space drugs on Twitter. Worth a read" (thread). The thread explains that complex cancer drugs such as PROTACs form uneven crystals on Earth. As a result, they dissolve poorly and their bioavailability is hard to predict. Varda's bet is that growing the crystals in microgravity yields purer, more stable drugs. Note that Delian cofounded Varda.
Shorter mentions
- Keith Rabois called a post by @jhong "good." The post argues that people who haven't run companies underrate how much the team matters compared with "some imagined structural moat."
- Guillermo Rauch recommended @DilumSanjaya's Arduino explainer app to educators, homeschoolers and learners. His point is that a coding agent can now produce engaging explainers.
- Patrick Collison called Ben Thompson "for many years ... the best analyst in tech" when announcing the Stratechery–Stripe partnership.
- Sarah Guo pointed again to the No Priors episode with Greg Brockman, citing his view that security spending should "vastly outstrip" software spending. The episode's topics include cybersecurity in the AI age.
- Tim praises Seti’s podcast about couples’ money problems and highlights its question about investing 10× more in one or two important areas as a prompt to rethink priorities.
- For durable management principles, he recommends reading old books and singles out Peter Drucker’s The Effective Executive as short and dense in tactics and principles, with outdated details easy to update.
- For breaking out of overthinking and inaction, Tim says James Clear’s Atomic Habits does a good job explaining how small actions accumulate; he advocates starting with small daily actions to rebuild self-trust.
- Kevin Rose recommended Michael Singer’s The Surrender Experiment to Tim as a counterexample to overengineering life; Tim says he enjoyed it and found it helpful, though it raised questions for him about total versus selective surrender.
- For communication, Tim names Crucial Conversations as a book to read; for couples, he says to choose Terry Real’s audio book Fierce Intimacy if picking one, praising Real’s clarity and saying the audio format may suit people with more time to listen than read.
- On back pain, Tim says John Sarno’s Healing Back Pain has helped friends and considers Sarno’s work valuable, but rejects its near-universal psychosomatic explanation and one-size-fits-all approach. He also calls Back Mechanic by Stuart McIll a great book and says he is a fan.
- Tim recommends Dr. Kevin Tracey’s The Great Nerve for learning more about the vagus nerve and calls his podcast conversation with Michael Leaven a great resource. He also directs listeners interested in metabolic psychiatry to his interview with Chris Palmer of Harvard.
- Tim passes along On Grief and Grieving: Finding the Meaning of Grief through the Five Stages of Loss by Elizabeth Kubler Ross and David Kesler, saying it was recommended most by people he trusts and had helped friends after bereavement; he notes he had not read it recently.
- For affordable fitness, he points new parents to Jersey and Angela Gregor’s The Happy Body and Coach Chris Summer’s gymnastic strength training (GST), describing both as requiring little equipment; he says combining them with acroyoga put him in the best shape of his life in decades.
- Tim stands by the slow-carb diet in The 4-Hour Body and points readers to a free blog post on how to lose 100 pounds on the diet, which he calls simple and reliable with a high likelihood of success.
Elon Musk linked an @KanekoaTheGreat post featuring a four-minute compilation of 27 James Talarico clips from 2019–2026, introducing it as “Talarico, the creep, in his OWN words.” Musk also told followers to “Forward this to your friends in Texas”; the linked post urged sending it to 10 Texas voters and telling them to vote that day, giving the recommendation a political-distribution context.
Reid Hoffman strongly supported OpenAI’s release of its mathematical proofs, saying the results show frontier models doing impressive work and reveal capabilities beyond what people see in everyday AI products .
Tobi Lütke praised David (“DHH” in the exchange)’s recent speech on “AI delirium” as “a fantastic one.” Lütke’s interpretation—while noting DHH might not agree—was that it describes people moving beyond the p(doom) debate and focusing on using AI to bring previously impossible things into the world.
Sarah Guo praised Alex Konrad’s Upstarts post on Watchtower Labs, thanking him for examining the military operations and logistics challenge and Watchtower’s effort to address it; the linked post reports the startup’s $35M launch from stealth and an interview with CEO Akhil Iyer.
Tobi Lütke praised DHH’s recent speech about “AI delirium” as “a fantastic one.” Lütke said this was his interpretation, which DHH might not share: “AI delirium” describes people who have moved beyond the p(doom) debate and are focused on bringing into the world what AI makes possible.
Jensen Huang pointed attendees to a website for an article about NVIDIA and Anthropic’s partnership, calling it “really great”; the excerpt does not identify the article or provide its URL.
Jensen Huang directed viewers to the website for what he called a “really great write-up” about NVIDIA and Anthropic partnering to create cloud OpenShell agents; the passage does not identify the write-up’s author or provide its URL.
Patrick O'Shaughnessy invited viewers to enjoy his second conversation with Christina Cacioppo, Vanta cofounder and CEO. The interview covers software economics, Vanta's AI rebuild, talent, AI hype and security, and trust; no separate reason for recommending it is given beyond those topics. Link to the conversation.
Aravind Srinivas encouraged users to try the Graphify skill by @graphify, available in Perplexity Computer; it turns a codebase into an interactive map of its files, functions, and dependencies . He said each user was getting 5,000 credits to try it; Perplexity’s announcement specified that Graphify tasks were free through October 31, capped at 5,000 credits per user .
Delian (@zebulgar) recommended Wajahat’s post on Varda and “space drugs,” calling it the first explanation he had seen that understood the subject and saying it was “Worth a read.” Wajahat argues that Varda aims to use undisturbed microgravity crystal growth for complex drugs such as PROTACs, potentially producing purer drugs with more predictable bioavailability and greater stability.
Sarah Guo (@saranormous) spotlighted a No Priors episode, “What are the Post-AGI Rules of the Game in Cyber and Science?”, featuring OpenAI co-founder/president @gdb; listed topics include cybersecurity in the AI age and how AI changes cybersecurity. In a later cybersecurity thread, she linked the episode as context for @gdb’s view that security spending should “vastly outstrip” software spending.
Patrick Collison praised Ben as “for many years” the best analyst in tech and said he was looking forward to the Stratechery–Stripe partnership, broadly endorsing Ben’s analysis rather than recommending a specific piece . He linked to the partnership page: https://stratechery.com/2026/presented-by-stripe/
Sarah Guo calls “can’t” not being in their vocabulary “the best reference you can get on founders” and says she is excited. The post includes an attached video.
Keith Rabois endorsed @jhong’s X post, calling it “good”; the post argues that a company’s team can matter more than an imagined structural moat, using VCs’ basketball-team analogy.
Sarah Guo shared “How Harvey is Building a Frontier Marketing Org”, by Gabe Pereyra and Shawn Farsai; her caption frames it as “Figma MCP 🤝 Harvey marketing.” The article’s example: Harvey used agents to make a launch film entirely in Figma and to create tailored product films for target accounts, with agent-written scripts and ElevenLabs narration.
Aaron Levie points readers to @fiveoutofnine’s roundup of cybersecurity vulnerabilities, breaches, and outages as an example of why cybersecurity will generate AI-related jobs and become a major enterprise investment area. He says agentic and autonomous systems will still need people to direct agents, review findings, patch systems, and respond to incidents.
Sarah Guo (@saranormous) called a piece by Alex Konrad at @UpstartsMediaCo “great,” recommending it for more background on Watchtower’s product and first contracts; the recommendation appeared in a thread about Conviction co-leading Watchtower’s $35 million seed round.
Tobi shared a linked post with “Of course it’s net negative,” endorsing its conclusion about the proposed California wealth tax . The post discusses Stanford academics Jaros and Rauh’s paper, The Net Present Value of the California Billionaire Tax Act, and reports that its simulations found a negative net present value 85% of the time; it links the NBER paper .
The sponsor blurb says Arianna Huffington praised Nabeela Elsayed’s Breaking Burnout but gives no reason for the praise. Book
How Harvey is Building a Frontier Marketing Org
How Harvey is Building a Frontier Marketing Org

Recently, @figma asked us what Harvey was building with its MCP server. This is the long answer.
Between the MCP and better frontier models, agents now run whole design projects end to end across our brand, product marketing, and design work. Here’s what that looks like, and the systems underneath it.
Building a Launch Film With Agents
We made this launch film for Harvey Artifacts entirely with agents, inside of Figma. The agent built every scene as a Figma Motion frame, keyed every camera move, UI state, and line of type on its timeline, and exported the cuts straight from the file.
It all started with a storyboard, which is to a film what a PRD is to code: it sits between the creative brief and production, so every shot, line, and transition is decided up front. That forces the model, and us, to be deliberate about every choice.
The agent reworks the board after each pass and asks which notes should become rules. For music, we marked where the beat should drop and it retimed every scene to the track. When the ending fell flat, it cut and EQ’d the track, added sound design, and laid out four or five ways the outro could land.

Tailoring a Demo for Every Account
Anyone can get one flashy demo out of an agent. The hard part is scale, so we picked a sales problem: a tailored product film for each of the hundreds of accounts we want to win. Our first prompts came back half decent, and 15 to 20 minutes of back-and-forth got each one into shape. But that wouldn’t scale.
So we set out to build a strong foundation in Figma, in two parts. Agents crawled the whole app, clicked every button, and rebuilt all key screens as components, which became Product Map. Then, in Figma Motion, they worked out the grammar of a product film, from camera pushes to screen handoffs to a steady click rhythm, which became Motion Primitives. Both are covered below.
With both in place, we worked with the agent to build a menu of 44 company-agnostic demo flows, each built around one moment buyers ask to see, like a playbook redline in Word. For each account, agents use Enterpret’s MCP to pull together Gong calls, Salesforce win-loss reports, and Slack. An agent then writes the script, with a fictional company on screen, and voices it as Ashleigh Whittaker, our narrator, whose voice we built in ElevenLabs from ten minutes of her footage.

Making Hundreds Overnight
An engine of about 750 lines of JavaScript turns each script into a film inside Figma. Each film is a short list of beats, and the engine handles the camera, transitions, captions, the pointer’s click rhythm, and the end card. Then it lays out contact sheets of the finished film and flags clipped text, small type, and stray pointers, so the obvious mistakes are caught before anyone watches.
Lots of small things pop up at scale. Like only one agent can work in a file at a time, so agents queuing in one shared file pushed calls to two minutes and timed out. We built a bus to run them in parallel: three lane files, each with its own builder agent, plus a coordinator for export, mixing, and delivery. One batch of 36 films took 69 minutes, and when we hit our daily Figma call cap, overnight agents scripted and voiced the next 97.
Hundreds of films also mean a lot of people to coordinate. So everything runs through one tracker, a Claude artifact that every agent and person reads and writes. Each film has a row with its account, story, status, and links, and agents update their own rows as they build, export, and deliver.

Agentic Brand Guidelines
We think about our Figma the way engineers think about a codebase. Our Agentic Brand Guidelines are the top-level AGENTS.md, and everything above runs on them. Each project is a folder with its own AGENTS.md page that adds only what it needs, and Agentic Brand Guidelines tell agents how to set one up with its own context and changelog. It’s progressive disclosure for creative work: an agent reads the brand, then the project, then only what its task needs.
We started with markdown files, and for design tokens they work fine. Everyone has their colors and type in a file somewhere. Markdown breaks one step up, when you try to explain how to make a chart for a blog post or structure a long-form PDF guide. A picture really is worth a thousand words: most of a design lives in the relationships between elements, and words describe those badly. So our Agentic Brand Guidelines became a Figma library, from foundations up to templates with worked examples, where agents see those relationships instead of reading about them.

Layout is where Figma beat HTML canvases for us. Designers work with grids the web was never built for: modular, baseline, compound, and hierarchical grids. We tried to get agents there in HTML, even building a layout engine on top of Pretext (opens in new tab), but browsers set type in line boxes padded by the font’s ascent and descent, so a headline’s cap height never sits exactly on a grid line. Figma trims text to cap height and baseline, hangs punctuation, and holds each grid as a layout agents apply.
Product Map and Motion Primitives
Our product is designed in a lot of places now. Product designers are more productive than ever, but some work is vibe coded and some lives in Figma, spread across many files. We needed one source of truth an agent could reference and maintain. Product Map is that source, published to our Figma org as a library, so any video that needs the product pulls it from there instead of redrawing it.

Motion was the other gap. Films kept animating the same pieces from scratch, and with little differences. So we built Motion Primitives, a library of animated parts that’s also published to our Figma org, down to product moves like redlines arriving in a document. We made sure the product pieces are exactly right, and a fix to one component reaches every film that uses it.

ElevenLabs
Every line of voiceover in our films comes from @ElevenLabs. Agents call it through its MCP, and our voice library already holds clones of our narrator and of teammates from their interviews, each built from a short clip of footage. We’re excited to try more of what ElevenLabs offers to make our content even better.

Figma Weave
Figma Weave does for images and video what ElevenLabs does for voice. It’s a canvas for chaining image and video models into workflows, and it’s now part of Figma, so our agents reach it through the same MCP they design with. When a frame calls for a photo, a texture, or a shot, the agent runs a model or one of our saved Weave workflows and places the result straight in the Figma file. Our Agentic Brand Guidelines spell out the art direction, so that the agent can easily pick whichever model and workflow in Weave gets the best result.

Multiplayer Creative Work With Spectre
Spectre is our internal agent platform. Engineers use it to run coding agents and our sales team uses it as a data agent. Shawn and Joey from our dev tools team pointed it at the brand repo and Figma files, and it became a creative agent that works in a Slack channel. Someone tags it on a request, and it does the work in Figma, then replies with what it made, what it changed, and which decisions to check.
We think this is what multiplayer design looks like now. It used to mean several designers in one file. Now the people who own the work in product, legal, sales, and marketing give direction in the Slack thread where they already work, while an agent makes the changes in Figma and everyone watches the work move.

Telling Harvey’s Story at Scale
So that’s the long answer to Figma’s question. In the last 14 days alone, Shawn has used 15.6B tokens across Figma, ElevenLabs, and Weave. All signs point to this only growing. Designers are starting to scale their token use the way engineers have over the last four years, and MCP is what makes that possible.
That’s letting us be far more ambitious with our marketing, from a launch film made entirely in Figma to a tailored product video for every account we want to win. As the models get better, so does what a team our size can make.
Authors: @ShawnFarsai and @gabepereyra
Sarah Guo shared “How Harvey is Building a Frontier Marketing Org”, by Gabe Pereyra and Shawn Farsai; her caption frames it as “Figma MCP 🤝 Harvey marketing.” The article’s example: Harvey used agents to make a launch film entirely in Figma and to create tailored product films for target accounts, with agent-written scripts and ElevenLabs narration.