"Service as Software" is Silicon Valley's hottest buzzword right now.
Everyone's talking about SaaS becoming service providers, but no one's explaining HOW. The answer? After 6 months of research and 100s of startup conversations, we have the answer: Systems of Agents.
We're looking at a $4.6T opportunity.
I'm sharing something unique we've been making at Google (w/ UNC). We are releasing our work on a new class of interactive experiences that we call generative infinite games, essentially video games where the game mechanics and graphics are fully subsumed by generative models 🧵
For the past 25 years, application software startups have had a singular focus: increasing company and employee productivity. AI enables a paradigm shift: Rather than sell software to improve an end-user's productivity, consider what it would look like to sell *the work itself*.
When you sell the work, instead of selling a 15% productivity improvement, you sell a 95% improvement. Incumbents are not only advantaged in selling software, they are stuck there. You don't need to play their game. Post: https://t.co/29BkdGIMC3
The new Realtime API with web crawling is mind-blowing!
Talk in realtime with any website. Powered by the OpenAI Realtime API and @firecrawl 🔥
Check it out:
If your design's visual fidelity outstrips its conceptual fidelity, you project a false sense of certainty to collaborators and risk constraining conversations about the product to decoration and style.
Excited to share our latest research on Contextual Retrieval - a technique that reduces incorrect chunk retrieval rates by up to 67%.
When combined with prompt caching, it may be one of the best techniques there is for implementing retrieval in RAG apps.
Let me explain:
Our latest course on LLM prompt evaluations is out.
Evals ensure your prompts are production-ready as you're able to quickly catch edge cases and zero in on exactly where your prompts need work.
Let's walk through what the course covers:
🚀 Optimizing LLMs for Accuracy 🚀
Boost LLM accuracy with these steps:
1️⃣ Start with prompt engineering for a solid baseline.
2️⃣ Add few-shot examples for consistency.
3️⃣ Use RAG for dynamic context.
4️⃣ Fine-tune for domain-specific tasks.
#llms#openai#langchain
I needed a dithering effect for some project photos, but none of the available tools were cutting it. So, I built my own! It only took a couple of hours, and I’m pretty proud of the result
I remain confused by the "GenAI is a dud" arguments.
We have controlled experiments showing 20%-40% gains in real work using GenAI. Adoption rates are the fastest in history. There is value.
I think what folks may mean is that "companies haven't captured the value created yet."
A small request: If you think this work is interesting, consider sharing it with someone you know who is "not a computer person". https://t.co/uAg40FbcFM
Why Github Actually Won
"I can boil it down to exactly two reasons that happened to resonate with each other at the perfect frequency.
- GitHub started at the right time
- GitHub had good taste"
https://t.co/ghBIychuxr
Dynamicland's new website, documenting ten years of work, will be released tomorrow at https://t.co/TxY7OKAuyg.
In the meantime, this video about Dynamicland's precursor gives some backstory and motivations. https://t.co/cvCDnKVIcW
Introducing the new CLI.
Install anything from anywhere—add components, themes, hooks, functions, animations, and generated code to your apps.
This marks a major step forward in distributing code that both you and your LLMs can access and use.
Let’s take a look.
We've raised $60M from Andreessen Horowitz, Jeff Dean, John Schulman, Noam Brown, and the founders of Stripe and Github.
Cursor has become recognized as the best way to code with AI, powered by an ensemble of custom and frontier models, delightful editing, and petabyte-scale infrastructure.
Our mission is to create a magical tool that will one day write all the world's software. Join us!