2026-03-19

RT Seth Bannon Israeli terrorism is out of control. The US shouldn't support a country that allows this. CNN International: Israeli settlers have increasingly used violence against Palestinians in a bid to drive them from their homes in the occupied West Bank. But sexual assault appears to be a new weapon in these settlers’ arsenal of intimidation, pointing to a troubling new level of violence.

RT Chris Rickard I'll be speaking @ AI Engineer Melbourne on "Legacy Software + Agentic Discovery". I’ll be sharing practical lessons from large-scale reverse-engineering projects: - recovering intent from code - where humans still matter most - what high-quality spec generation might look like ... and when I say large, I mean 12M+ LOC. ... and when I say legacy, I mean 25+ years old. --- 3-4 June 2026, Federation Square Melbourne & Online. Sharing the stage with some rad humans including @swyx, @GeoffreyHuntley, @jeremyphoward and stacks more. 🔗 Tix && discount link: https://webdirections.org/register/?eventName=aieng26&eventTitle=AI+Engineer+Melbourne+2026&selectedTicket=aieng26conference&promoCode=chrisraieng26conference%2Cchrisraieng26streaming Huge thanks @aiDotEngineer & @johnallsopp

RT Jake The demand for software (and compute) is unlimited 15k, gapping +40% w/w Jake: 12k new users per day Gapping up 20% w/w

RT Lucas Crespo 📧 Google Stitch is interesting but let’s be honest about what it is. Spell check didn’t make anyone a better writer. It just made bad writing less frequent. That’s what AI design tools do. They clean up the mess. They don’t create the vision. You can prompt claude code, gemini, or any of them and they will all give you something competent. Not bad. Just not interesting. When everyone has access to “pretty good,” pretty good stops working. It becomes noise. The baseline shifts and very rapidly the only stuff that cuts through the noise is the stuff that has a point of view. That takes an eye and imagination and I promise you no prompt box is ever gonna give you that. In summary, good designers are definitely not cooked. Stitch by Google: Meet the new Stitch, your vibe design partner. Here are 5 major upgrades to help you create, iterate and collaborate: 🎨 AI-Native Canvas 🧠 Smarter Design Agent 🎙️ Voice ⚡️ Instant Prototypes 📐 Design Systems and DESIGN.md Rolling out now. Details and product walkthrough

Screw the haters, I am long @figma

So cool! Michelle Lim: This is the era of the one-person marketing team. Launch dozens of high-performing landing pages from Claude and Clay. Using @tryflint’s MCP and API.

Never could I have imagined that I'll be writing blog posts via the terminal one day

Had meetings and a dinner with 20+ enterprise AI and IT leaders today. Lots of interesting conversations around the state of AI in large enterprises, especially regulated businesses. Here are some of general trends: * Agents are clearly the big thing. Enterprises moving from talking about chatbots to agents, though we’re still very early. Coding is still the dominant agentic use-case being adopted thus far, with other categories of across knowledge work starting to emerge. Lots of agentic work moving from pilots and PoCs into production, and some enterprises had lots of active live use-cases. * Agentic use-cases span every part of a business, from back office operations to client facing experiences from sales to customer onboarding workflows. General feeling is that agentic workflows will hit every part of an organization, often with biggest focus on delivering better for customers, getting better insights and intelligence from data and documents, speeding up high ROI workflows with agents, and so on. Very limited discussion on pure cost cutting. * Data and AI governance still remain core challenges. Getting data and content into a spot that agents can securely and easily operate on remains a huge task for more organizations. Years of data management fragmentation that wasn’t a problem now is an issue for enterprises looking to adopt agents. And governing what agents can do with data in a workflow still a major topic. * Identity emerging as a big topic. Can the agent have access to everything you have? In a world of dozens of agents working on behalf, potentially too much data exposure and scope for the agents. How do we manage agents with partitioned level of access to your information? * Lots of emerging questions on how we will budget for tokens across use-cases and teams. Companies don’t want to constrain use-cases, but equally need to be mindful of ultimate token budgets. This is going to become a bigger part of OpEx over time, and probably won’t make sense to be considered an IT budget anymore. Likely needs to be factored into the rest of operating expenses. * Interoperability is key. Every enterprise is deploying multiple AI systems right now, and it’s unlikely that there’s going to be a single platform to rule them all. Customers are getting savvier on how to handle agent interoperability, and this will be one of the biggest drivers of an AI stack going forward. Lots more takeaways than just this, but needless to say the momentum is building but equally enterprises are acutely aware of the change management and work ahead. Lots of opportunity right now.

This is what a modern slot machine looks like. Except 100x more addictive 🥹

RT Garry Tan OK Codex is GOAT at finding bugs and finding plan errors

Let's go Matt! 🚀 Matt Jayson: 📣 Huge update!! We raised $9.5M. And we’re introducing @multiply_ads - the first hybrid AI media agency for B2B. We help high-ambition B2B companies find new customers faster. Marketing is at the center of every great company. But advertising is broken. You launch

RT Akshay Kothari Figma’s best days are still ahead. I’m long the company, and super long @zoink!

trust @reckless to put the seo slop at the end haahahahhahaa hero and yes at work we have one of those fancy wifi printers with screens and doodads and it has worked exactly 1 time out of the ~10 times ive tried to print something. castform5: @uncledoomer Verge printer recommendation article always relevant https://www.theverge.com/23642073/best-printer-2023-brother-laser-wi-fi-its-fine

RT Garvit is building Kulfi We showed up at @spc_india’s consumer demo day - and now are packing our bags for SF ✈️ @kulfiapp is now backed by @southpkcommons Super excited to learn from the ecosystem and come back ready to build at a completely different level @adityaag @rsanghvi @prateekmehta42 @ankitcc @aadhar @naushad_arsalan

RT Dylan Itzikowitz Our friend Kevin Ryan has been instrumental in building the NYC tech ecosystem & is still shaping it Last night we united the next generation of builders in NYC together NYC’s talent community is dense, spans industries, and keeps growing. @southpkcommons @alley_corp

RT Gabriel Bianconi Excited to demo TensorZero Autopilot, our automated AI engineer that analyzes LLM observability data, optimizes prompts and models, sets up evals, and runs A/B tests.

Start building your first app for free at https://ai.studio/build Logan Kilpatrick: Introducing the all new vibe coding experience in @GoogleAIStudio, feating: - One click database support - Sign in with Google support - A new coding agent powered by Antigravity - Multiplayer + backend app support and so much more coming soon! https://x.com/GoogleAIStudio/status/2034654985850659149?s=20

RT Siddharth Codex's ability to reverse engineer private macOS frameworks is uncanny. Just added an HDR toggle to DisplayBuddy - so many users asked for this!

RT Federico Cassano a lot went into this model. it was fun! i hope people enjoy it. Cursor: Composer 2 is now available in Cursor.

RT Keshav Jindal Disappointed. the design sucks. it's not mobile firendly at all. generated words don't persist. @Replit did so much better Google AI Studio: vibe coding in AI Studio just got a major upgrade 🚀 • multiplayer: build real-time games & tools • real services: connect live data • persistent builds: close the tab, it keeps working • pro UI: shadcn, Framer Motion & npm support we can't wait to see what you build!

RT Paul Graham Another advantage of focusing on growth rate rather than absolute numbers is that it makes it easier to switch to a new variant of the product if you discover one. It makes it easier to see tails that will eventually wag the dog.

RT Marcus Williams Sharing some of the work I’ve been doing at OpenAI: we now monitor 99.9% of internal coding traffic for misalignment using our most powerful models, reviewing full trajectories to catch suspicious behavior, escalate serious cases quickly, and strengthen our safeguards over time.

RT Joe Robison Just surf the frontier models Dan Shipper 📧: How to never lose your job to AI: Just surf the models. Frontier models outclass humans at any form of knowledge that can be written down. But people who use frontier models in their field of expertise generate new, tacit, situational expertise that the models don't yet

RT Replit ⠕ Have you tried the Replit Design Canvas yet? 👀 1. Generate multiple design variants 2. Convert your design into app 3. Fine tune designs with design controls 4. Sketch your idea & make it real 5. Collaborate with your team on designs

Frontier intelligence. Fraction of cost. Fast. Composer 2. Cursor: We were able to significantly improve the model quality and cost to serve. These quality improvements come from our first continued pretraining run, providing a far stronger base to scale our reinforcement learning.

Perhaps I should get married again so that the media has a more recent man they can reference any time they mention me or my work.

RT Rick Heitzmann IRL is an important megatrend and @poshvip_ and @avanteprice are making it easier to build events and find your people! Avante Price: I'm excited to announce Posh’s $37M Series B, led by @FirstMarkCap. 6 years ago, people told us a better events platform was a solved problem. 8m users and $350M in GMV later, we're just getting started. For more on our vision and what's next: http://posh.vip/vision

RT Billy Howell Built a working Nano Banana wrapper for $5 in @Replit tokens Took an hour Agent 4 is really sweet

RT Dan McAteer Felix Rieseberg, Anthropic's Claude Cowork lead, shares an incredible insight on @latentspacepod. This insight is something that you'll only get from an insider engineer: "Is the right thing to invest in these scaffolding corrections? Or is it to give the model many capabilities, try to make those safe, then just wait a second for the next model to drop?" That's where Anthropic engineers who work directly with Claude see the puck going. You need to do the same. Assume that, whatever you are building, the model will continue to get exponentially better, and your custom scaffolding becomes irrelevant.

every is such a special place to work because i randomly bump into beautiful things like this in figma while looking for something else

RT Amjad Masad The most magical thing about this is once when you’re done infinitely exploring designs you can just click “Build” and you get a production app! Replit ⠕: Have you tried the Replit Design Canvas yet? 👀 1. Generate multiple design variants 2. Convert your design into app 3. Fine tune designs with design controls 4. Sketch your idea & make it real 5. Collaborate with your team on designs

The most magical thing about this is once when you’re done infinitely exploring designs you can just click “Build” and you get a production app! Replit ⠕: Have you tried the Replit Design Canvas yet? 👀 1. Generate multiple design variants 2. Convert your design into app 3. Fine tune designs with design controls 4. Sketch your idea & make it real 5. Collaborate with your team on designs

RT basil Replit is the only vibe coding company that I can use for anything serious Fairly shocking on how good the “design videos” with Replit is Amjad Masad: The most magical thing about this is once when you’re done infinitely exploring designs you can just click “Build” and you get a production app!

we just published the definitive guide to writing good prose with your AI read it on @every: https://every.to/guides/ai-style-guide?source=post_button

RT Katie Parrott I've been waxing rhapsodic about the value of an AI writing style guide for long enough that it felt rude not to write up a guide on how to make one. So we did. Now on @every https://every.to/guides/how-to-build-an-ai-style-guide

RT morgan — google deprioritises their chrome agent after openclaw goes vertical, browser agent usage “essentially amounts to a rounding error” Max Zeff: New: Google has moved some staffers off its Project Mariner browser agent to higher priority projects, WIRED has learned. Browser agents were the hottest thing in tech last year, but Claude Code, OpenClaw, and the coding agent craze are reshaping the industry's priorities

RT Every 📧 just surf the models Dan Shipper 📧: How to never lose your job to AI: Just surf the models. Frontier models outclass humans at any form of knowledge that can be written down. But people who use frontier models in their field of expertise generate new, tacit, situational expertise that the models don't yet

come be my friend please Reem Ateyeh: I'm hiring someone to join my team at Anthropic to lead Claude Code comms. This is not a role for someone who wants to run an old playbook. You'll need to be a Claude Code super user, understand developers and dev tools, and have great taste. You'll work hard, learn a lot, and

i asked composer 2 to optimize my production QA process and pitted it against gpt-5.4 composer 2's response won (as judged by both 5.4 and opus 4.6):

RT Ryan Delve, a YC-backed compliance startup that raised $32 million, has been accused of systematically faking SOC 2, ISO 27001, HIPAA, and GDPR compliance reports for hundreds of clients. According to a detailed Substack investigation by DeepDelver, a leaked Google spreadsheet containing links to hundreds of confidential draft audit reports revealed that Delve generates auditor conclusions before any auditor reviews evidence, uses the same template across 99.8% of reports, and relies on Indian certification mills operating through empty US shells instead of the "US-based CPA firms" they advertise. Here's the breakdown: > 493 out of 494 leaked SOC 2 reports allegedly contain identical boilerplate text, including the same grammatical errors and nonsensical sentences, with only a company name, logo, org chart, and signature swapped in > Auditor conclusions and test procedures are reportedly pre-written in draft reports before clients even provide their company description, which would violate AICPA independence rules requiring auditors to independently design tests and form conclusions > All 259 Type II reports claim zero security incidents, zero personnel changes, zero customer terminations, and zero cyber incidents during the observation period, with identical "unable to test" conclusions across every client > Delve's "US-based auditors" are actually Accorp and Gradient, described as Indian certification mills operating through US shell entities. 99%+ of clients reportedly went through one of these two firms over the past 6 months > The platform allegedly publishes fully populated trust pages claiming vulnerability scanning, pentesting, and data recovery simulations before any compliance work has been done > Delve pre-fabricates board meeting minutes, risk assessments, security incident simulations, and employee evidence that clients can adopt with a single click, according to the author > Most "integrations" are just containers for manual screenshots with no actual API connections. The author describes the platform as a "SOC 2 template pack with a thin SaaS wrapper" > When the leak was exposed, CEO Karun Kaushik emailed clients calling the allegations "falsified claims" from an "AI-generated email" and stated no sensitive data was accessed, while the reports themselves contained private signatures and confidential architecture diagrams > Companies relying on these reports could face criminal liability under HIPAA and fines up to 4% of global revenue under GDPR for compliance violations they believed were resolved > When clients threaten to leave, Delve reportedly pairs them with an external vCISO for manual off-platform work, which the author argues proves their own platform can't deliver real compliance > Delve's sales price dropped from $15,000 to $6,000 with ISO 27001 and a penetration test thrown in when a client mentioned considering a competitor erin griffith: A detailed and brutal look at the tactics of buzzy AI compliance startup Delve "Delve built a machine designed to make clients complicit without their knowledge, to manufacture plausible deniability while producing exactly the opposite." https://substack.com/home/post/p-191342187

Code is the foundation of all knowledge work. Another proof point right here: Anton Osika – eu/acc: Introducing Lovable for more general tasks. Lovable has always been for building apps. Today it also becomes your data scientist, your business analyst, your deck builder, and your marketing assistant. This is a big step toward what Lovable is becoming: a general-purpose

RT Thariq We just released Claude Code channels, which allows you to control your Claude Code session through select MCPs, starting with Telegram and Discord. Use this to message Claude Code directly from your phone.

RT Brandon Gell tell me this isnt the future. agents working with other agents to build repeatable skills you can share across your team

RT Harold Hunt It used to be that only the most cracked developers could debug heap snapshots and find the root cause of a memory leak. When Node.js unit tests are slow, most developers just accepted it. But with Agentic Engineering you can make a skill out of it... and Codex is shockingly good at analyzing the root cause of high memory usage when guided to collect and analyze heapsnapshot deltas. Feel free to steal this: https://github.com/openclaw/openclaw/pull/50654

Forbes 30 under 30 remains undefeated (fraud)

AI Learned My Email Habits

No transcript available

Benedict Evans: OpenAI’s Moat Problem & the Future of Software

No transcript available