Follow builders, not influencers.
TL;DR
Woodward said Gemini’s turning into a test-prep machine, Albert called Claude Code the whole workspace, and Cat Wu shipped a desktop control center with parallel sessions and review tools. Rauch also argued agent builders need elastic Postgres, not vibes.
Gemini now has NEET practice tests for India’s biggest medical exam, with more subjects and countries on the way. It’s a pretty clear signal that Google wants Gemini to be useful for high-stakes studying, not just chat.
He says the new Claude Code desktop app ships with a bunch of feature and performance upgrades, and now he barely needs other apps — or even a terminal — for most work. That’s a pretty strong signal Anthropic is pushing Claude from chatbot into full-on work surface.
He says enterprise AI agents won’t kill the forward-deployed engineer — they’ll make that role more important. Selling agents is really selling a workflow, which means vendors need domain expertise, system wiring, and change management or adoption will stall.
Claude Code on desktop is now the place to juggle multiple sessions across local and cloud, with git status, pinned runs, and drag-and-drop layouts. She also says Routines can kick off Claude Code on a schedule, from GitHub events, or via other APIs — basically turning it into an automation layer, not just a chat box.
She says the real work on Tab Out wasn’t adding features — it was deleting them. What started as an LLM-heavy tab manager got stripped down into a pure Chrome extension with fully local storage, because the useful part was simply seeing tabs clearly and closing them fast. Her takeaway: Claude will happily overbuild unless you keep it on a leash.
He says simple workflows will get vibe-coded away, but real systems still take time, persistence, and ugly details. The Slack notification system is his example: you can prototype fast, but quality and depth still have to be earned.
He says the stuff he used to hack together as one-off conversion apps is now just a skill, which is the real shift: AI is making app-building repeatable instead of bespoke. He also flagged a practical Replit update for Indian builders — you can now build and pay with Razorpay, which removes a real bit of friction.
He says anyone building an agent coding platform should pair app generation with a highly elastic, Postgres-compatible database. It’s a quiet product nudge, but the point is clear: if AI is going to spin up apps fast, the backend has to keep up — and Vercel’s DSQL is the example he points to.
He jokes that AI is giving every VC the same investment advice, which is exactly the problem: the more everyone leans on the same tools, the more herd thinking sneaks in. It’s a pointed reminder from the FirstMark VC and MAD Landscape author that “independent thinking” is getting harder to prove, not easier.
He says people overreact to Google and Meta’s misses because their execution unit is massive. The right lens, especially for a product leader at Linear, is to count the wins that come with that scale instead of fixating on the obvious failures.
He says GBrain is MIT-licensed and built in about 12 days, powered by his GStack coding toolkit and a deep OpenClaw addiction. The pitch: his own instance is already over 17,000 pages and “knows” him, so this is less demo and more a living personal knowledge system.
They introduced Sparkle v4, an @every app that acts like a desktop janitor: it scans your filesystem, suggests an organization system, and deep-cleans junk like old installers, duplicates, and screenshots with permission. The bigger pitch is that it keeps your Mac tidy in the background on a schedule, so chaos doesn’t creep back in.
He says AI output isn’t the finish line — it’s raw material. In his chat with Dylan, the point is that agents can generate endless options, but you still have to judge, shape, and push past the average with taste. That’s a useful frame for anyone building with Claude Code or similar tools: the model speeds exploration, but the human still decides what’s worth keeping.
Claude Code desktop adds parallel sessions and built-in review tools
Lead: Claude Code’s redesigned desktop app is built for parallel agent workflows, adding a new sidebar for managing multiple sessions plus integrated tools for reviewing and shipping work without leaving the app.
Numbers:
So What: The update turns Claude Code into more of an orchestration workspace than a single-chat interface: you can run sessions across repos, branch into side chats, edit files, run tests, inspect diffs, and preview outputs in one place. The app also matches CLI plugins, supports drag-and-drop layouts, and streams responses as Claude generates them. Anthropic says the redesign is meant for “how agentic coding actually feels now: many things in flight, and you in the orchestrator seat.” For builders, the practical takeaway is simple: if you’re already using Claude Code, you can now keep more of the coding loop inside the desktop app instead of bouncing between terminal, editor, and browser.
Marketing now has to persuade agents, not just people
The Takeaway: Marketing is shifting from ranking for humans to being legible, useful, and quotable to agents.
James Cadwallader, cofounder and CEO of ProFound, thinks marketing has entered a new operating system. His company helps brands understand how they appear inside AI search and agentic answers across models like ChatGPT, Claude, Gemini, and Grok. His core argument is blunt: “the biggest platform shift in the history of marketing” is underway, because the consumer is being replaced by an agent making the buying journey on their behalf.
That changes the job. In the old world, marketers optimized for human attention and the top few blue links. Now, Cadwallader says, agents use “orders of magnitude wider” surface area, sometimes pulling from dozens of pages for a single query. That means brands need content that is not just keyword-rich, but structurally legible to models. For software especially, he pushes beyond legibility into usability and interoperability: can the agent actually troubleshoot, compare, and recommend your product?
He’s also skeptical of the idea that AI content is automatically junk. His view is that the real constraint is originality. If models already know everything on the internet, marketers have to give them something they don’t know. Or, as he puts it, humans are now “a fleshy API between reality and the Internet.”
Rauch said the moat moved from code to the code factory, while Levie argued every team now needed an agent wrangler. Cursor leaned into customizable multi-agent views, Replit added region controls, and No Priors backed Periodic Labs’ bet that AI could learn atoms by running experiments.
Amjad Masad said Apple’s 50th has turned into a PR disaster, while Aaron Levie argued agents would create more work, not cut jobs. Rauch pushed engineers into the customer hot seat, and Claude warned teams to harden security fast.
Thariq said Claude Code now handles TurboTax pain, while Rauch called microVM sandboxes the new compute layer. Aditya Agarwal pushed memory over loops, and Levie argued AI won’t shrink law—it’ll inflate it.
Claude pushed into Word with tracked edits, and Claude Code moved planning to the web with auto mode approvals. Garry Tan called agents the Altair BASIC era, while Aaron Levie warned software without a real API gets left behind.
Karpathy said free ChatGPT lagged while frontier coding models didn’t. Albert pushed cheap-to-smart escalation, Rauch said cloud infra went agent-native, and OpenAI’s next leap looked like autonomy—not chat.
Woodward gave Gemini a second brain with Notebooks, while Anthropic shipped Managed Agents to move Claude from prompt to production. Rauch called the web AI’s native OS, and Levie, Masad, and Shipper all bet agents will do the work, not the people.
Albert teased Anthropic’s Mythos Preview, Cat Wu juiced Claude Code’s CLI tricks, and Peter Steinberger patched CodexBar with 2 providers plus billing fixes. Levie said agents are eating knowledge work, while Nikunj Kothari preached retention over launch hype.
Levie said agents won’t erase work, just push it up a layer; Yang argued they’ll shrink teams, not ambition. Garry Tan flagged an unpatched file leak in Claude’s coding env, while Kothari called Anthropic’s revenue ramp absurdly fast.
Rauch said v0 now builds physics, not just UI, while Karpathy noted GitHub Gists have weirdly good comments. Levie argued AI efficiency creates more work, not less, and Tan called open source’s golden age.
Karpathy pushed “your data, your files, your AI.” Levie argued context beat raw model IQ in enterprise AI. Garry Tan said GStack kept shipping security fixes fast, while No Priors spotlighted Periodic Labs’ bet on atoms, not just text.
Claude plugged into Microsoft 365 everywhere, Swyx said Devin one-shot blog-to-code, and Peter Steinberger called out GitHub’s API as still not built for agents. Aaron Levie hit the context wall, while Garry Tan shipped a DX review tool from his own stack.
Claude landed computer use on Windows, Karpathy argued LLMs should build your wiki, and Amjad Masad pushed Replit deeper into enterprise sales. Peter Yang said Cursor 3 got out of the agent’s way, while Peter Steinberger warned AI slop was flooding kernel security with real bugs.
Steinberger called plan mode training wheels, while Thariq gave Claude Code a mouse-friendly renderer and Cat Wu showed sessions jumping phone-to-laptop. Masad framed Replit as an OS for agents, Rauch said Vercel signups compounded fast, and Anthropic’s infra tweaks swung coding scores by 6 points.
Levie said AI productivity hit the enterprise risk wall, while Weil argued proofs got cleaner, not just better. Agarwal floated public source code as the new prod debugging, and Data Driven NYC claimed one founder could run a company if agents handled the layers below.
Karpathy warned unpinned deps can turn one hack into mass pwnage, while Rauch and Levie said agents still need human guardrails and redesigned workflows. Meanwhile Claude Code got enterprise auto mode, Replit added built-in monetization, and Swyx spotted “Sign in with ChatGPT” already live.
Andrej Karpathy highlighted how LLMs can argue any side, suggesting we use it as a feature. Guillermo Rauch finally shipped his dream text layout, bringing his vision to life. Meanwhile, Amjad Masad claimed AI is democratizing app building and elevating top engineers.
Andrej Karpathy suggested leveraging LLMs' ability to argue any side as a feature. Guillermo Rauch turned text layout dreams into reality with Vercel's latest feature. Meanwhile, Amjad Masad claimed AI is democratizing app building, liberating top engineers for bigger challenges.