SID.ai (YC S23)

"End of RAG" or "Jevons Paradox"? Some say long-context language models (LCLMs) will kill RAG. Even RAG's inventor himself. Their logic? Why fish for information when you can flood the prompt with an ocean of data? We think they overlook a fundamental truth about human nature and technological progress… Jevons Paradox – historically, when a resource's efficiency increased, its consumption skyrocketed: Electricity got cheaper, so we ended up using more. With any finite context window size, there's always an application that benefits from having even more information: Alas you need some mechanism – retrieval – to extend past the window. Even if we were to reach infinite context length, RAG remains relevant due to: • Efficiency: Processing queries with billions of tokens is costly and slow. Retrieval lets us pick just the important ones. • Data Access: LCLMs still need bridges to various data sources. • Better Reasoning: RAG can cut through noise and handle complex tasks like temporal reasoning, multi-step logic, and hierarchical data far better than LCLMs. The future isn't RAG vs. Long Context — it's RAG + Long Context. What's your take? Will RAG become obsolete, or will it evolve alongside LCLMs? 🤔 More on this topic with Raza Habib & my co-founder Max Rumpf on the Humanloop "High Agency" podcast. 👇

What if San Francisco's secret isn't its network, but the crazy? We've all heard the hype around SF: land of unicorns, bottomless VC pockets, game-changing networks. But building a company on both sides of the Atlantic has shown me a defining quality of the Bay Area that often goes unnoticed. Growing up in Switzerland is a masterclass in societal perfection. The downside? Conformity. People call you "special" as an insult. Even those that claim to value innovation secretly despise what's truly different. The nail that sticks out... gets hammered down (especially girls with big dreams). In SF, that same nail gets celebrated. This is the city’s unfair advantage: – Not the startup ecosystem. – Not the capital. – But the freedom to be weird. The radical acceptance of the unconventional. It's in the city's DNA – from the Gold Rush of '49 to the Summer of Love in '67. SF has long been a haven for those who want more than the status quo. Why it's so important? You can't build a generation-defining company by coloring inside the lines. You need to cultivate the space to be strange, to think different, to be contrarian, to see what others can't – or won't. San Francisco doesn't just allow you to be weird. It dares you to be weirder.

RAG Complexity ≠ RAG Accuracy Many trendy RAG features fail to significantly improve real-world performance... I've been building RAG for 2+ years now. And every week brings a shiny new approach, research paper, Twitter thread. It's tempting to cram every cutting-edge technique into your system. But more buttons on a coffee machine don't necessarily mean tastier coffee - just more ways to mess up your brew. Just like more complexity in a RAG system doesn't necessarily guarantee better performance... Our philosophy? Ruthless pragmatism! Every new technique must prove its worth. Does it improve the end-to-end performance of our customers' use-cases in the real world? No metric increase – no inclusion. No exceptions! This might seem obvious. Yet it's surprising how many chase complexity for complexity's sake. We've tested countless RAG components rigorously – tree-based chunking, sliding window with overlap, dense vector search, query expansion, multi-vector indexing – the list does on... Some work wonders – the really do! – others... not so much. The kicker? It all depends on your specific use case. Happy to give you some pointers if you're thinking about implementing a specific feature at the moment! Link to the full conversation of Raza Habib & Max Rumpf on Humanloop "High Agency" podcast in the comments.

The Importance of Being Frugal with Human Tokens Max Rumpf wrote an excellent article on "Amdahl's Argument for AI", suggesting that humans-in-the-loop might be the biggest limiting factor in boosting productivity through AI. My key takeaway: Copilots are a temporary phenomenon – agent-like approaches are here to stay. Why? Speedup of ChatGPT & GitHub's copilot: The speed at which humans can process and respond (about 1-3 tokens per second) sets a fundamental limit on productivity gains. For example, applications that require a human completion for every LLM completion can achieve a maximum speedup of ~2x, even when LLMs become 10x faster. Speedup of current agentic approaches: Cognition's Devon demonstrates the potential for greater productivity gains by requiring human completion only every ~10 iterations. At the moment, this approach leads to ~2.9x gain over ChatGPT with current model speed. Speedup of future AI agents: The real game-changer are AI agents in the 100-1000x range, i.e. those that only require human input every 100-1000 iterations. Granted, it will take some time to get there – but we're already seeing many approaches: Like human-less code execution from E2B, browsing from Browserbase or the context engine from SID.ai (YC S23). Right now, developers are frugal with LLM tokens due to their cost. However, I agree with Max that the most important thing to be frugal with are human tokens (both input and output) – because they will define the overall productivity speedup your application can provide.

Sam Altman’s advice at yesterday’s Turing event: Right now, AI startups fall into two distinct categories: Those (implicitly) betting that AI models will stay at the current capability level and those betting the models will get better. The former build things to address the current shortcomings. The latter want the models to get smarter, easier to prompt, more accurate, and faster – because that will make their own products better. Predictably, Sam recommended you to bet on the latter: Models will keep getting smarter and we won’t hit a wall on model intelligence “in the next millenium.” Otherwise, he said (jokingly), the “OpenAI killed my startup” meme will inevitably become your future. What’s your take on the development of model capabilities?

🥧 Pi Day seems like the perfect time to show you SID.ai (YC S23)'s brand new office at 3141 (π ≈ 3.141). We're thrilled to be part of the vibrant Mission district – sharing our block with incredible neighbours like MindsDB and OpenAI. If you ever need a change of scenery or find yourself nearby, please don't hesitate to drop in. We'd be thrilled to invite you in for a coffee or have you co-work with us for the day. Seriously, I mean it – come by anytime! Here's to many intense days and nights within these new four walls! 📦🦥

Excited to announce a small update for #AgentGPT v1.0 ✨ Featuring: - A freshly redesigned theme 🚀 - Updated agents with a focus on information retrieval through search 🔎 - SID.ai (YC S23) for personalized context and queries 🦥 - Increased usage limits for all users 📈 - Automatic name generation 🤖 Try for free at https://lnkd.in/exWykEmD

Excited to announce a small update for #AgentGPT v1.0 ✨ Featuring: - A freshly redesigned theme 🚀 - Updated agents with a focus on information retrieval through search 🔎 - SID.ai (YC S23) for personalized context and queries 🦥 - Increased usage limits for all users 📈 - Automatic name generation 🤖 Try for free at https://lnkd.in/exWykEmD

How to get into Y Combinator – 101 Almost 1,000 people reached out after my last post! So I took the time to write down all the things that helped us get into YC. The good news: They’re the same things you need to build a great, successful company. 🏀 Take every shot you have – aka apply! You’re neither too late nor too early. Some people get accepted with just an idea (or less), while others have thousands in ARR. If you haven’t applied yet – do so ASAP. An application is never “wasted” – it’s always better to apply to both this batch and then to all future ones if you don’t get in the first time around. The great thing about the YC application: The questions they ask help you reflect on and strengthen your idea. 🏎️ Write progress updates! Pretend it’s a race that starts the second you apply. Months may pass before YC reads your application. Push yourself to make as much progress as possible and update your YC application. The best part: The focus and efficiency will let you make progress you thought was impossible. And remember: The race never stops. ❤️🩹 Be resilient. If you’re committed to working on this, accept the fact that you’ll get rejected constantly. You’ll hear no’s from everyone: customers, investors, incumbents... Keep trying, again and again and again. Early startups are a matter of how many shots on goal you can take and how much you learn from the misses. 🩲 Be brief. Don’t waste words explaining to YC how large you think GenAI will be or why computers are great. Trust me, they know! Neither should your application be a buzzword bingo. Being brief helps with customers and investors, too. Customers won’t buy something they don’t understand. 🚀 Launch your product today, not tomorrow. If you’re live: keep shipping! We always tried to launch early – yet every time I look back, I think we could have gotten our product out the door even quicker. If you’re adding a dark mode, you’ve gone too far! Launch ASAP and get some users on it (and update the application). ✉️ Push sales, early and hard! If you’re working on a problem worth solving – you’d be surprised by how willing people are to give you money. The reverse is also true: Why build something that people won’t pay for? As Justin Kan famously said: “First-time founders are obsessed with the product. Second-time founders are obsessed with distribution.” 🤝 When you ask (YC alumni) for recommendations – make sure that the person knows you well. Recommendations from strangers won’t help your case. The same holds for other recs, intros, reference checks – e.g. never take an intro from an investor who has not invested in you! If you do end up scoring an interview, let me know. I’ll gladly give you more pointers as to what to expect and how to succeed. I might even have time to jump on a quick trial interview. In the meantime, my fingers are crossed! 🤞 Make the next couple of weeks count – so that you make enough progress to write your application updates about! 😉

To all the women that have not yet applied to Y Combinator's winter batch - and to everyone who knows a great woman that should: Don't worry that the application deadline has passed. You can still apply late! Reach out to me or intro them to me – I'll gladly take a look at your application or give you a prep talk! 🥰 We at SID.ai (YC S23) wrote our application just 24h before the deadline – leading to the (so far) most educational and amazing summer of my life! So take your shot! Please share this with your network so we can reach more awesome women! Also a big shoutout to our group partners Diana Hu and Surbhi Sarna for taking extra good care of us girls during the summer batch! 🫶