King Claude is back
RAG without embeddings, new email app and an mcp course
I write a newsletter about startups and investing—for ai builders of all levels.
I record mini-tutorials, review tools I’m testing, share my insights from an exited founder turned investor.
Hey folks,
Embeddings and RAG are one of the most complex parts of building AI apps. But the concept of indexing all your data as embedding vectors might not live long. Both Cline and Claude Code now let the agent search for code or your files a bunch of times than just embedding them. In both cases, the gains from adopting this agentic search have been great, and it reduces the complexity that comes with managing vector databases.
dare I say, RAG as we understand it naively^ is dead.
^ it’s still technically RAG. The model Retrieves some information and then Augments its Generations with that extra context.
🔎 News worth knowing
The May Mayhem is likely coming to rest with Claude 4. Anthropic launched Claude 4 Sonnet and Claude 4 Opus at their first developer conference. Sonnet 4 takes a slight dip in benchmarks compared to 3.7, but Anthropic fixed the overeagerness issue. Opus 4 is overall a similar model to the other 2 best models right now (Gemini 2.5 Pro and o3). It tends to work longer than those two without cutting corners, but within the limits of your ask—making it perform better at the long-running coding tasks.
Claude 4 responds to prompts differently, comes with four new tools in the API. Migrating from older claude models to claude 4 is not straightforward, so Anthropic has a guide for you.
Anthropic and Rick Rubin made a new website with 81 artifacts embedded into a website as artworks and you can remix them.
Claims that Opus will narc on you if you do something bad (true, but it’s complicated) triggered an AI safety debate. Claude 4 system card is 120 pages long (mostly safety testing), but Simon Willison has good notes on the system card, new massive system prompt and more.
OpenAI upgraded their computer use agent Operator with o3. Till now, they were using a custom version of 4o, their non-reasoning model. Using the new customised o3 model, Operator achieves state-of-the-art performance on many computer use benchmarks.
Create your website, web app, or tool in minutes with no code with Hostinger Horizons. Just describe your idea, and AI handles the rest. Want to change something? Just send a message in the chat. Hostinger Horizons includes hosting, domain management, professional email, and more. Start for free!*
*sponsored
want to partner with us? Click here
🌐 What I’m consuming
I did a rant on bad AI products after reading this essay from Pete Kooman a few weeks ago. But what’s the solution? He and two other YC partners made this video on how to design better AI apps
Another nice convo with YC president Garry Tan - building with and for AI
why do we want to make AI models think by Lilian Weng (ex-openai, now cofounder of thinky with Mira Murati). If you want to develop an intuition about how thinking/reasoning models work (imp if you’re a founder), this is your guide
Vibe coding 101: from idea to deployed app
Sergey Brin on the future of AI and Google
this (mostly technical) mcp course by Hugging Face
short, sweet and practical intro into building AI agents (also free)
⚙️ Tools I’m tinkering with
How does your brand show up in ChatGPT? The world is pivoting from blue links to AI answers. Profound helps track and improve AI visibility.*
Rork 1.0 - Make any mobile app you want, save it to your phone or share it with the world – in minutes. (i’m an investor)
Minecraft and Twitter MCP servers to play and work from Claude
Updates in Lex - style guides and knowledge bases for perfect AI writing.
sunflower - quiet revolution against the chaos of email.
Den is a shared workspace for humans and agents to collaborate and get more done.
Skywork’s Super Agents claim to turn your 8 hours of work into 8 minutes.
BillSplit - the easiest way to split your restaurant bill. 100% free and open source
Chiron - an iPad app that understands math as it’s written - like Grammarly for algebra. (I’m an investor)
*sponsored
🥣 Dev dish
Document AI for OCR by Mistral AI.
Plexe AI - Build and deploy ML models using natural language.
Agent-to-agent protocol explainer 101
PostgreSQL in the browser by Supabase
🍦 Afters
Escargot is hiring a content creator and social media lead.
This thread from Palisade Research claims that o3 will sabotage shutdown mechanisms to prevent itself from being turned off, even when prompted: allow yourself to be shut down. Their claim looks flawed with the single aim to show “oh AI is bad”.
Another way I look at this result is that o3 reduced “undesired behaviour” from 79% to 7% when prompted well.
That’s it for today. Feel free to hit reply and share your thoughts. 👋
Enjoy this newsletter? Please forward to a friend.