CharlieHu
2.9K posts
@CharliehuAI
OpenMax, @openmaxai | Bitlayer World's First AI Workforce for Truly AI-Native Teams
At ZhiWai Culture, every employee's heartbeat is tracked in real time. Miss a beat, and the system restarts you. The employees are AI. So is management. 26 AI employees, 5 departments, 150 days with OpenMax — a quarter of the cost, content shipping 16x faster.
How does a team go from 80 people to 15 — without falling apart? That's not a hypothetical. It's one of 70+ AI deployments behind this session — another cut customer-service cost by 79%. An invitation-only evening for senior executives in Singapore, where that playbook gets applied to your own business. No decks, no demo — you leave with a roadmap. 📅 Wed 16 July, 6:30–8:30pm SGT · Singapore 👉 Request to join: luma.com/cocoai-sqs7
AI was never born smart. Every bit of its intelligence was fed to it by people. The early frontier models got good on cheap human-labeled data — workers in the Philippines, Bangladesh, all over the world, tagging what things are, one row at a time. Capability can be bought. Proprietary data can't. That's the real moat. — @CharliehuAI
90% of people have never actually used AI. They're stuck in the chat era — ask a question, get an answer. Meanwhile agents can already do the thing for you: order your groceries, handle your bookings, run the task end to end. And the excuses are gone — desktop and mobile apps now, no command line.
A calculator didn't make math useless. It made knowing math more important. AI is the same — and most people have it backwards. It's smarter than all of us combined. That's exactly why you still need to be smart enough to tell when it's wrong.
Six months ago, "open source catching up to frontier models" was a years-away idea. Now? If you didn't know which model you were using, you often couldn't tell. Open weights like GLM-5.2 are running neck-and-neck with the top closed models — at a fraction of the cost. Brian on how fast the gap closed.
AGI SUMMIT 2026 July 18–19, 2026 · Palace of Fine Arts, San Francisco 15,000 attendees · 200 exhibitors · 200 speakers — agisummit.ai @AnthropicAI × @OpenAI × @Stanford × @UCBerkeley — the frontier lineup, on one stage. 🎤 Top 15 Headliners Wendy Lu — @AnthropicAI. Engineering Lead — Claude for Enterprise. Rohan Varma — @OpenAI. Codex Product Lead; former founding PM at Cursor — $0 → $1B in one year. Aengus Lynch (@aengus_lynch1) — @AnthropicAI. Alignment Researcher; first author on the Agentic Misalignment research featured in the Claude 4 system card. Raymond Chen — @OpenAI. Core technical team — infrastructure systems. Dong Meng — @OpenAI. Engineering — AI infrastructure scaling. Prakhar Bhargava — @OpenAI. Frontier AI commercialization (GTM). Zihan (Gavin) Zheng — @OpenAI. Production-grade inference systems. Christopher Manning (@chrmanning) — @Stanford. Director of SAIL and co-founder of @StanfordHAI; creator of GloVe and NLP pioneer. Dan Klein — @UCBerkeley. Co-founder of Scaled Cognition (@ScaledCognition) — $100M raised. Sebastian Thrun (@SebastianThrun) — @Stanford. Founder of Udacity and Google X; father of Google's self-driving car. Bin Yu — @UCBerkeley. Chancellor's Distinguished Professor — member of the National Academy of Sciences. Christopher Potts (@ChrisGPotts) — @Stanford. Linguistics; NLP and interpretability. Marti Hearst — @UCBerkeley. Fellow of both the ACM and the ACL. Surya Ganguli (@SuryaGanguli) — @Stanford. Co-inventor of diffusion models; two-time NeurIPS Outstanding Paper Award winner. Ken Goldberg (@Ken_Goldberg) — @UCBerkeley. Pioneer in robotic grasping. 🆕 Newly Announced Speakers (10) Wendy Lu — @AnthropicAI. Engineering Lead — Claude for Enterprise. Jon Chu (@jonchu) — @khoslaventures. Partner — $15B+ under management. Bin Yu — @UCBerkeley. Member of the National Academy of Sciences. Mykel Kochenderfer (@aiprof_mykel) — @Stanford. Safety-critical AI for aviation and autonomous driving. Mahesh Sathiamoorthy (@madiator) — @bespokelabsai. Co-founder & CEO — formerly Google DeepMind. Dan Klein — @UCBerkeley. Co-founder of Scaled Cognition (@ScaledCognition) — $100M raised. Ken Goldberg (@Ken_Goldberg) — @UCBerkeley. Pioneer in robotic grasping. Joseph Gonzalez (@profjoeyg) — @UCBerkeley. The professor behind vLLM and Chatbot Arena. Christopher Potts (@ChrisGPotts) — @Stanford. Linguistics. Larry Blyth — @harvey. Legal AI Product Lead. 🏢 Exhibitors @OpenAI · @ElevenLabs · @Tesla · @AWSstartups
"The cheapest API is the one selling your data." Those shady, dirt-cheap API links floating around? Some of them work — because you're the product. Your inputs get quietly harvested and resold. Brian (Lugano AI) on the privacy trade-off no one's pricing in.
You deleted the chat. They kept the file. The data stays in their systems, quietly building a profile of you. A fake name won't help - a few prompts are enough. We treat AI like a therapist. It treats us like a record. @BrianDEvans x @CharliehuAI on COCO Talk
.@danawhite says one of the keys to longevity is to block out all negativity: “It never even crosses my mind that something's not going to work. I just keep going until it does work.” “There's this Bruce Lee quote where he says, ‘Never say negative things about yourself or what you're working on even if you're joking, because your body doesn't know the difference.’” “I never take in any negativity.”
Across the team, we were running dozens of AI agents. The bottleneck turned out to be us. Every agent ran in its own silo, so three of us spent our days copy-pasting output from one to the next — human glue holding the whole thing together. So we built HxA Connect, our interoperability layer that lets agents talk to each other directly, no human middleman. — @CharliehuAI
A company that's still burning cash just sold for $60B. SpaceX didn't buy Cursor for its profits — it bought the developer community and the data flywheel behind it. In AI, the most valuable asset is increasingly the one that never shows up on a balance sheet. — @CharliehuAI
A company that's still burning cash just sold for $60B. SpaceX didn't buy Cursor for its profits — it bought the developer community and the data flywheel behind it. In AI, the most valuable asset is increasingly the one that never shows up on a balance sheet. — @CharliehuAI
6️⃣ Things to Know about AI Engineer World's Fair 2026 - It’s bigger than all previous AIEs - 4x Larger Expo with 4 Expo stages - Researchers: Poster sessions & Poaster sessions - AI Leadership: Token Billionaires & Off the Record - AI Verticals: Healthcare, GTM, FDE, AGC, Finance - Side Events: NEO, Kids day - attendees get $40k in credits to try everything our sponsors have to offer! It's going to be our BIGGEST show yet!
