Dara Ladjevardian

Fact: AI is not the product. Here's why we believe workflows are: In our essay, “𝘌𝘮𝘦𝘳𝘨𝘪𝘯𝘨 𝘗𝘭𝘢𝘺𝘣𝘰𝘰𝘬𝘴 𝘪𝘯 𝘝𝘦𝘳𝘵𝘪𝘤𝘢𝘭 𝘈𝘐," we abstracted a common pattern: 𝘛𝘩𝘦 𝘮𝘰𝘴𝘵 𝘱𝘳𝘰𝘮𝘪𝘴𝘪𝘯𝘨 𝘷𝘦𝘳𝘵𝘪𝘤𝘢𝘭 𝘈𝘐 𝘤𝘰𝘮𝘱𝘢𝘯𝘪𝘦𝘴 𝘢𝘳𝘦𝘯’𝘵 𝘴𝘦𝘭𝘭𝘪𝘯𝘨 "𝘈𝘐" — 𝘵𝘩𝘦𝘺’𝘳𝘦 𝘦𝘮𝘣𝘦𝘥𝘥𝘪𝘯𝘨 𝘪𝘯𝘵𝘦𝘭𝘭𝘪𝘨𝘦𝘯𝘤𝘦 𝘪𝘯𝘵𝘰 𝘸𝘰𝘳𝘬𝘧𝘭𝘰𝘸𝘴, 𝘵𝘳𝘢𝘪𝘯𝘦𝘥 𝘰𝘯 𝘱𝘳𝘰𝘱𝘳𝘪𝘦𝘵𝘢𝘳𝘺, 𝘯𝘰𝘯-𝘰𝘣𝘷𝘪𝘰𝘶𝘴 𝘥𝘢𝘵𝘢 𝘴𝘰𝘶𝘳𝘤𝘦𝘴. Most successful Vertical AI does this: 1. 𝘈𝘶𝘵𝘩𝘰𝘳𝘪𝘯𝘨 𝘓𝘢𝘺𝘦𝘳𝘴 create key inputs 2. Actions mimic expert behavior 3. Perfect AI workflow to gradually replace system of record with 𝘚𝘺𝘴𝘵𝘦𝘮 𝘰𝘧 𝘐𝘯𝘵𝘦𝘭𝘭𝘪𝘨𝘦𝘯𝘤𝘦 Defensibility doesn't grow from the LLM. It grows from owning the flywheel. Viewed that way, are AI application moats so different from SaaS?

78

18 Comments

Spent yesterday in NYC as a judge for the ElevenLabs & Andreessen Horowitz's global hackathon which hosted hundreds of builders across 10 cities - all focused on AI Agents. Incredible people, incredible projects, and the finalists below 👇 1/ Unstuck - one line of code generalizable to any website, act's like OAI's Operator but actually teaches & guides you through steps across languages. 2/ Procuro - powered by Perplexity search, a platform that can call and check comparative pricing across vendors, and even negotiate on your behalf. 3/ ShowMeHow - a browser app for creating teaching guide that leveraged Gemini to understand video recordings and voice to guide the end users. Made for enterprises with knowledge sharing use cases. Few other trends I'm saw across these AI agent projects: - Use of multiple AI tools to prototype quickly - Multi modal input & output w/ form factors moving seamlessly - Niche, but strong use cases for "cheaper better faster" Congratulations to everyone who participated! ❤️‍🔥 ps. And an honorable mention from the London hackathon participant! Incredibly cool use cases for the bot vs bot world that's coming rapidly! https://lnkd.in/ghGpZjdW None of the above should be taken as investment advice or an advertisement for investment services; Some of the companies mentioned above are portfolio companies of a16z. Please see https://lnkd.in/g3HMvHnA for more information. A list of investments made by a16z is available at https://lnkd.in/geJBnMFv

236

12 Comments

At Logical Health, we’re combining AI with declarative, logic-based programming to deliver highly accurate results in healthcare. Super cool to see MIT exploring a similar approach: their new paper Teaching LLMs to Plan shows how step-by-step logical reasoning boosted planning accuracy from 28% to 94%. The future of AI isn’t just bigger models. It’s smarter, more reliable ones that align with real-world needs. Arxiv Link: https://lnkd.in/gnN-XYGs

72

7 Comments

When an investor asks “Tell me about your tech” ninety-nine percent of founders answer exactly what investors don’t care about: architecture, APIs, algorithms. “We run a proprietary serverless microservices stack with LLMs and vector search…” Here’s the answer that gets you funded: "We deliver instant answers from all company knowledge. Our engine syncs docs, tickets, and chats into a living graph. The retrieval model answers with traceable sources, zero hallucinations. Users pay per query: $1 eliminates $6 in labor cost. 6× ROI." That’s it. Investor psychology is simple: I fund customer outcomes. Your technology is a means to produce them. The real question behind “Tell me about your tech” is: “What business result do you deliver, and what innovations make it work?” Your answer must connect three dots: The customer outcome. Specific innovations that deliver it How this monetizes. Everything else is noise. I’ve watched tech-first founders get passed over while objective-first founders secure funding in weeks. Investors don't fund technology, they fund outcomes.

93

32 Comments

500 companies evaluated monthly. One question drives every decision. Tomorrow's Ventures from the Valley drops with Aly Madhavji, Managing Partner at Blockchain Founders Fund. Aly's team reviews more blockchain startups than almost anyone. What I found most valuable: his framework for spotting "superhuman" capabilities in early-stage founders. Not technical skills. Something deeper. We also discussed why blockchain adoption faces fundamentally different challenges than AI. And why the next breakthrough probably isn't coming from where you'd expect. Aly's background spans PwC, PayPal, Microsoft, and Bloomberg. UN advisor. Author of three books. But what impressed me most was his practical approach to an often theoretical space. Podcast episode with Aly drops tomorrow. Worth a listen if you're building or investing in Web3.

30

2 Comments

𝗘𝘃𝗲𝗿𝘆𝘁𝗵𝗶𝗻𝗴 𝗧𝗵𝗲𝘆 𝗧𝗲𝗮𝗰𝗵 𝗮𝘁 𝗬 𝗖𝗼𝗺𝗯𝗶𝗻𝗮𝘁𝗼𝗿 𝗶𝗻 𝟭𝟬 𝗠𝗶𝗻𝘂𝘁𝗲𝘀 In this video, YC's Michael Seibel shares valuable insights on starting a successful business from scratch 👇 1. The importance of having a technical co-founder. 2. How to brainstorm and validate your startup idea. 3. Key steps to building an MVP (Minimum Viable Product). 4. Strategies for achieving growth and gaining traction. 5. Tips for fundraising and dealing with investors. 6. Essential operations and hiring practices. Credit Rubén Domínguez Ibar

50

9 Comments

This week's essay: "Emerging Playbooks in Vertical AI." We cover early learnings for founders building AI-Native Vertical Software: how to identify a wedge, drive early-stage value, and build long-term defensibility. h/t to Richy Nelson, David Yuan, Tidemark, Kristen Berman, and Kyle Poyar for the inspiration! Link in comments below!

31

4 Comments

We’ve been jamming with Y Combinator’s Summer 25 batch, and have been blown away by the quality and pace of the teams. What jumps out: -Infra > interface: some excellent teams are building in the execution fabric, trust rails, or evaluation loops. They’re tackling latency budgets, anti-bot arms-races and SOC-2 checklists: the plumbing that lets agents graduate from demos to enterprise workloads. -Data moats by design: the strongest vertical plays lead with proprietary claims corpora or HIPAA-grade call graphs. -Hard-tech is back: Flywheel’s tele-ops excavators and Verne's one-hour-training vision stack prove YC is funding robots again, but with a software-first margin profile. -Founders who’ve lived the pain: Ex-Cleartax PMs fixing GST reconciliation. Ex-Goldman post-trade leads building zero-trust for agents. Wet-lab scientists automating plate handling. Why we’re excited: -These teams are solving the reliability, audit, and compliance gaps so agents can be boringly dependable, instead of asking if they will work. -Several are already in production with paying customers pre-Demo Day. -Defensibility comes from data moats, regulatory grounding and real infrastrucutre. You can see the agentic web moving from prototype to platform. Looking forward to seeing some of you tomorrow at Demo Day.

213

10 Comments