boldstart ventures
4.5K posts
@Boldstartvc
partnering from Inception with bold technical founders building the autonomous world, lead investor for pre-product founders @edsim @etdurbin @ellenchisa
🔥 up to announce @boldstartvc Fund VII $250M to back bold technical founders building the autonomous enterprise. From Inception. Before the world believes. It always starts with an idea that feels insane… until it isn’t. 🎥👇🧵
I spoke to IBM CEO Arvind Krishna about the warning today and here’s what he told me: “A few large capex deals paused near the end of the quarter. It’s not widespread. It’s mainframe and the software associated with it….Mythos is making people pause to say, wait how much do I need to spend on cyber? They’re pausing on new deals until they know…We don’t see our software being disrupted by ai at all.”
🚀 Accelerating SGLang HiCache with Netpreme X-Mem™ MPU: new blog from @netpreme on plugging a TB/s memory tier into HiCache's hierarchical KV cache! In multi-turn coding sessions (Claude Code on SWE-bench), prefix-cache hit rates average ~98%. At this point, KV bandwidth is the bottleneck, not compute. Swapping Host DRAM for X-Mem™ as the offload tier: ✅ Up to 6.7× lower TTFT ✅ 33–50% higher TPS/user under load ✅ 30% higher system throughput ✅ Drop-in via HiCache's tiering interface
We're opening the waitlist for our Monetization Gateway, which will allow you to charge for any web page, dataset, API, or MCP tool behind Cloudflare. The charges will settle in stablecoins over the x402 open protocol. cfl.re/4eUFdt6
Satya said it well: “There is no societal permission for an AI future that eats the intelligence of the companies it’s deployed inside.” Which FDE would you rather have? Forward Deployed Engineering or Forward Deployed Extraction …of your data, workflows, and institutional knowledge. 👏🏻 @Microsoft
Your AI coding agent keeps making the same mistakes. Not because the model is bad. Because nobody has time to teach it. That's what Tessl Agent is built to solve. Instead of waiting for someone to update prompts, write new skills, or automate recurring tasks, Tessl Agent watches how your team works. It learns from PRs, coding agent sessions, and issue trackers, then opens PRs with improvements and automates the repetitive work that keeps getting pushed to "later." It's an agent that quietly improves your existing agents and workflows over time, so your team spends less time fixing the same problems and more time shipping. Tessl Agent is now in open beta. Read more about it here: tessl.co/cuf Or try Tessl Agent today: tessl.co/hf1
This post is for software infra founders who are feeling a bit beaten up at moment…
Palantir CEO Alex Karp says enterprises want to "own the means of production" instead of "transferring their alpha" to OpenAI or Anthropic: "Why are [LLMs] charging for tokens if it's so valuable?" "If it was so valuable—let's say I can make you a billion dollars tomorrow. Wouldn't I say, 'I'll make you a billion dollars, and I want 30%?'" "Look at our financials. The reason why everyone is chillaxing with bad financials and growth while losing money, is the client refuses to pay the true cost." "The two places that actually make money—profit, free cash flow—are our application layer called ontology, and compute." "We can get the frontier application to be exactly the same as a frontier model without the risk of transferring the alpha of your business to another." Via @CNBC
That's one of the coolest robotics demos! 😮💨 @GeneralistAI showed GEN-1 handling box folding and screw packing during @AutomateShow. The boxes are cardboard with real variability: creasing, deformation, different configurations. GEN-1 retries when things go wrong. It adapts mid-task. This is a jump from GEN-0, shown 3 months ago during GTC. GEN-0 handled rigid, predictable boxes. GEN-1 handles deformable ones with chaotic behavior. The scaling is straightforward. GEN-1 trained on 500,000 hours of proprietary, real-world dexterous manipulation data. Scaling laws are working. More high-quality real-world data leads to better physics understanding, better retry behavior, better generalization. And the time to get a working demo is shrinking drastically. They're not even at scale yet. They're just starting to see what happens when you pre-train on massive amounts of actual robot data. ~~ ♻️ Join the weekly robotics newsletter, and never miss any news → ziegler.substack.com
Salesforce just paid $3.6 billion for the technology I invented in 1997. Sit with that for a second. FIN. Formerly Intercom. AI agents that resolve customer queries through chat buttons on websites. That's it. That's the thing. Chat on a website, deployed at scale, sold for $3.6 billion to the world's largest CRM company. I know this world. I lived it for 28 years. I took LivePerson public on the back of this exact technology. Fast Company named us the number one AI company in the world in 2022, above OpenAI, because we had automated a billion conversations for the largest banks and telcos on the planet. And here's what I know from all of it. The chat button doesn't work. A fraction of customers ever click it. The rest call 1-800 numbers, sit on hold, explain their problem to three different people, and hang up more frustrated than when they started. We dressed it up in AI. We automated the resolution. We trained the models. And the fundamental experience stayed broken. I've wanted to kill 1-800 numbers since 2015. I said it from the Nasdaq opening bell. I meant it. I still mean it. But here's what I see now that I couldn't see then. The answer was never a better chat button on their website. The answer is you never go to their website at all. Your Individual AI already knows who you are. Your account. Your history. Your preferences. Your last three interactions with that brand. When you need something, you don't hunt for a chat window. You ask your AI. It has the connectors. It has the context. It reaches out to the brand on your behalf, fully armed with everything it needs to resolve the issue in one shot. The interface to customer support isn't a button on a company's website. The interface is you. That's the world I'm building at Uare. Not a smarter chat agent on their side of the wall. An intelligent, personal, context-rich AI on your side. One that represents you the way a great assistant represents an executive. Fully briefed. Fully authorized. Fully yours. $3.6 billion is a big number. It tells you this market matters. It also tells you the world is still solving the wrong problem. I'll spend every waking day finishing what I set out to do. To book end the whole story, my first investor from LivePerson, @edsim is the lead in Uare. We started this journey together and we will finish it together!
Artificial Intelligence (AI) bot traffic is growing 300% year over year. Publishers and content owners are forced to either block all bots or allow unrestricted access.
