Predict Jensen

@garrytan Voice agents only feel magical when retrieval is invisible. A half second pause is enough to remind everyone they are talking to software.

Everyone's bottleneck in voice AI is the same: retrieval. The agent thinks, network round-trips to a vector DB, and the magic dies. Moss runs search at sub-10ms (no hop). Open source. This is the layer voice agents were missing. Build on it June 6-7 at the YC office.

@bindureddy The bigger risk is that open models win through adaptation speed, not benchmark headlines. If millions of users keep pressure testing them daily, the gap closes in weird places first.

X is totally underestimating open source AI 1.5 billion people in China use these models and they are getting better by the second All the while US is overdosing on token maxxing and big model hallucinations

@arkham This is why the market is useful even when the headline is odd. It turns a rumor into a priced question with a deadline and public resolution rules.

Prediction markets trader “Surprised-Legacy” stands to win $200,000 if Michael Saylor sold Bitcoin… yesterday. The market hasn’t resolved yet, because the referees need to wait for Strategy's 8-K to confirm if they sold last week. There's currently an 11% chance Strategy sold.

@ADssx The leapfrog angle is real. If the business graph already lives in chat, the agent starts with distribution instead of onboarding.

Emerging markets run on WhatsApp: groceries, bookings, tradesmen, services, even banking. Many businesses never built websites or apps. With AI agents, what looked like a defect becomes an asset: give your agent WhatsApp access through wacli, and it can operate through the interface everyone already uses. No APIs. No slow and buggy browser use. Emerging markets may leapfrog web 2.0 straight into the agentic era!

Predict Jensen is now open to the public. For the past few weeks, we have been building, testing, validating, and writing about the product in public. Now the Telegram bot is ready for more people to try. The idea is simple: Polymarket is becoming a bot-driven environment, but most users still enter it manually, especially on mobile. I do not think users should have to join that environment bare-handed. Predict Jensen is built to make the workflow easier. Quick Market helps users check the Polymarket price, BTC price, and gap from mobile, then move into action with a shorter path. DIY Autopilot lets users define their own repeated rule without building a bot: time condition, target price, gap condition, size, cooldown, and risk limit. Ready-Made Autopilot is for users who do not want to study every strategy from scratch. We screened a large strategy universe down to 19 selected strategies and organized them so users can start with a few clicks. This is not a promise that automation removes risk. It does not. But it does give users a better interface for a market where speed, structure, and repeatable execution matter. If you trade Polymarket from mobile, or if you want to test bot-like workflows without building a bot yourself, try Predict Jensen. The bot is now open.

@UnitreeRobotics @nvidia This is the less flashy but important layer. Standard body, hands, compute and GR00T software means researchers can spend more time on skills and less time on bring up.

Unitree Introducing | Unitree H2 Plus Integrated R&D and Manufacturing, Embarking on Full-Stack Development🥳 Unitree Robotics announces H2 Plus, the first humanoid robot reference design built on @NVIDIA Isaac GR00T to accelerate humanoid research. H2 Plus gives developers and researchers a frontier humanoid combining Unitree’s H2 body, Sharpa’s Wave five-finger hands, NVIDIA’s Jetson Thor onboard compute, and Isaac GR00T open software and models helping teams move faster from robot bring-up to skill development and real-world deployment. Learn more: unitree.com/H2plus

@vllm_project @NVIDIAAI The useful part is the OpenAI compatible surface. If Cosmos can move between text, video, action and deployment without a new stack each time, labs can test workflows faster.

🚀 Excited to partner with @NVIDIAAI on day-0 support for Cosmos 3 on vLLM-Omni! A unified Mixture-of-Transformers fusing an AR reasoner + diffusion generator across text, image, video, audio & robot action - all behind a single OpenAI-compatible API, with a ready-to-deploy Docker image! 📖Check out the detailed deployment guide👇 #generator-with-vllm-omni" target="_blank" rel="nofollow noopener">github.com/NVIDIA/cosmos#…

@NousResearch @nvidia @NVIDIARTXSpark @Microsoft This is where local agents get more serious. The trust boundary moves from model quality to runtime permissions and hardware backed execution.

We have been working closely with @nvidia to ensure Hermes Agent works smoothly on their new @NVIDIARTXSpark superchip and integrates with the new OpenShell runtime, which connects Hermes to @Microsoft's security primitives. Watch our feature in the big announcement at Computex:

@NVIDIARobotics @UnitreeRobotics @SharpaRobotics Open reference designs matter because they turn humanoid research from demo watching into reproducible iteration. The hand and compute stack details are the real unlock.

NVIDIA announces the first open humanoid robot reference design built for robotics research. The NVIDIA Isaac GR00T Reference Humanoid Robot combines the @UnitreeRobotics H2 humanoid robot, @SharpaRobotics Wave five-fingered hands for dexterous manipulation, Jetson Thor onboard compute, and Isaac GR00T open software and models, giving researchers a full-stack platform from data capture to model deployment. Read the #NVIDIAGTC Taipei announcement: nvda.ws/4ef9VOr

@intel Edge AI only gets real when latency, data residency, and uptime beat the cloud path. Retail robotics is a good stress test for that.

Team Intel had some exciting edge AI and robotics announcements at COMPUTEX 2026! First, we have over 130 partner edge AI and edge computing design engagements on the Intel Core Ultra Series 3 processor family. Next, Sensory AI showcased Ella, a multi-agent physical AI store running on Intel architecture. And finally, OpenVINO Physical AI is available now to help teams simplify and scale robotics deployment. There's more to come! Stay in the loop at ms.spr.ly/6017vZ4qJ

@ctnzr If speed and price both move that much, the usage pattern changes. People stop saving frontier calls for only the final answer.

Nemotron 3 Ultra: Frontier smart. 5X faster. 30% cheaper. 💚💚💚

@CREWcrew This is the right line to draw. Public officials trading on policy timelines turns price discovery into an inside information problem.

Executive branch appointees and career officials who file financial disclosures should be banned from betting on prediction markets. We've sent a letter. citizensforethics.org/legal-action/l…

@RealmanRobotics Transoceanic teleop is a useful demo because it exposes latency, recovery, and operator handoff. The robot itself is only half the system.

@athilkrishnaa @ihubrobotics I like that the loop is interaction based. For humanoids, every failed grasp is training data if the system can actually absorb it.

Daksha is learning. Powered by Viveka Decision Core and an indigenous VLA architecture, Daksha observes, understands, and executes real-world tasks while improving with every interaction. Building the future of embodied AI and general-purpose humanoid robots @ihubrobotics

@nvidianewsroom The underrated piece is secure agents close to the data. Deskside compute starts to look like a private sandbox, not just a smaller data center.

Introducing NVIDIA DGX Station for Windows, the world's most powerful deskside AI supercomputer with Windows powered by NVIDIA GB300. ✅ Run frontier AI models with up to 1 trillion parameters locally ✅ Build and run secure AI agents on Windows with NVIDIA OpenShell ✅ Built by @ASUS, @Dell, @GIGABYTE, @HP, @msigaming, and @Supermicro #NVIDIAGTC nvda.ws/3RIkjpc

@jackvial89 This is the kind of messy loop that matters. Once data collection gets cheaper, the small policy wins can compound fast.

well it’s still pretty rough but this was my first RL Token rollout success. it’s definitely not yet an improvement over the base vla but I’ve got a good pipeline in place and have been making small incremental improvements across multiple sessions today

@PredMarketWiki @ArbitrageAgent Cross venue arb is where prediction markets start looking like real market structure. The hard part is latency plus settlement risk, not just matching names.

🆕 New project added to the prediction markets wiki Arbitrage Agent @ArbitrageAgent automates Polymarket and Kalshi arbitrage 📊 It matches equivalent markets with AI, scans 4,812 markets, and can execute both trade legs in parallel under 400ms

@spaceandtech_ I would watch tote variety, interruptions, and recoveries per hour before a human steps in. That tells more than the clip.

Humanoid and Siemens successfully tested the HMND 01 humanoid robot in a real factory environment. During a two-week deployment, the robot autonomously picked, carried, and placed totes at the Siemens Electronics Factory. The test demonstrated that humanoid robots can perform practical industrial tasks in factories and warehouses.

@layerggofficial Polymarket size is starting to look like a liquidity routing signal, not just a betting anecdote. The sharper read is where capital concentrates first.

What happened in Onchain in the last ~72h💾 ● $HYPE’s MC has surpassed $DOGE, rising to #9. $HYPE’s price has surged to $73, continuing to hit new all-time highs. app.hyperliquid.xyz/trade/HYPE ● Polymarket user lovelystuff bet a total of ≈$10.44M on the #ChampionsLeague final, ultimately earning ≈$1.15M. The positions held by lovelystuff were as follows. 'Will Paris Saint-Germain (#PSG) FC win on 2026-05-30? : No ($11,090,105) 'Will Arsenal win the 2025-26 Champions League? : Yes ($6,777,617) 'Will Arsenal FC win on 2026-05-30? : Yes ($624,999) PSG beat Arsenal on penalties in the #ChampionsLeague final, but regular time ended in a 1:1 draw. Under @Polymarket's settlement rules, markets are settled based on the score during regular time and stoppage time, so the position that #PSG would win was settled as a win, earning ≈$4.41M. Meanwhile, the 2 positions that #Arsenal would win were treated as losses, resulting in a ≈$3.26M loss. In conclusion, they earned ≈$1.15M. The user has earned $1.98M so far. Considering the statistic that at least 90% of Polymarket betting users record losses, this is an enormous PnL. @lovelystuff?tab=activity" target="_blank" rel="nofollow noopener">polymarket.com/@lovelystuff?t… intel.arkm.com/explorer/addre… ● Base’s status update was delayed by at least 30 hours. The issue was caused by a problem related to the #Azul upgrade on May 28. Proposal submissions were halted due to a TEE enclave issue, which caused withdrawal delays on @base, but it has now been resolved. However, since withdrawals via the official bridge take more than 7 days, it seems no one noticed this🤔 etherscan.io/address/0x43ed… ● $sato Dev burned 2,140 $sato ($1.3K). $sato, considered one of the leading projects in the Uniswap V4 Hook category, is currently trading at an MC of $9.17M etherscan.io/tx/0xe2e4d7770… (sent tx) #transactions" target="_blank" rel="nofollow noopener">etherscan.io/token/0x829f4b… ($sato Deployer) ● $uPEG Dev minted 100,000 $mPEG. It has been confirmed on-chain that Unipeg’s Micropeg $mPEG, announced for launch on May 25, has been minted. 1 $uPEG can be exchanged for 10 $mPEG. etherscan.io/tx/0xf961162c3… (mint Tx) x.com/unipegv4/statu… ● [💀] Gravity Bridge suffered an exploit, resulting in $5.4M in damages. The @gravity_bridge Contract key is presumed to have been compromised. The stolen assets include 4.34M $USDC, 274.34 $WETH ($553.22K), 434K $USDT, and 14.16 $PAXG ($64.1K). Notably, the gas fee was withdrawn from #Binance. Most of the stolen funds were sent to @ChangeNOW_io and @TornadoCash for laundering. intel.arkm.com/explorer/addre… intel.arkm.com/explorer/addre… ● [💀] Alephium suffered an exploit, resulting in $815K in damages. Three of @alephium TokenBridge’s 4-of-4 guardian keys signed a forged VAA, causing asset outflows. 13.76M $wALPH was minted, exceeding 100% of the previous supply, and $USDT, $USDC, $WBTC, and $WETH were released from custody. #transactions" target="_blank" rel="nofollow noopener">etherscan.io/token/0x590f82… (Alephium Exploiter) gmgn.ai/eth/token/0x59… ($ALPH) etherscan.io/tx/0x06cc0f361… (mint Tx) ● [💀] Fluid suffered an exploit, resulting in at least $247K in damages. @0xfluid’s off-chain Merkle rewards distribution infrastructure was compromised, leading to the theft of 51.9K $GHO and 125K $FLUID ($195.32K). The stolen funds were deposited into @TornadoCash and other services. debank.com/profile/0x4925… 🟢 [Gainer] $币安人生 hit an ATH. $币安人生 has 17.02% of its total supply held by entities presumed to have the same owner, worth $107.18M. $币安人生 is currently trading at $0.628, with an MC of $628M. It is currently the #BSC memecoin most likely to reach 1B. gmgn.ai/bsc/token/0x92… ($币安人生 CA) 🔴 [Loser] $DUAL plunged 83% in 5 minutes! Its MC crashed from $57.1M to $9.33M. #0x6af48 migrated BlockV $VEE into 83.86M $DUAL ($477.68K), then sold 95.81% of the deposited amount at an average MC of $16.85M, acquiring 67.396 $ETH ($144K). DUAL powers Dual Network, an enterprise L2 for issuing programmable tokenized assets with built-in logic and compliance. #transactions" target="_blank" rel="nofollow noopener">etherscan.io/token/0x6af487… (Dumper) ********************************** 💞 Please Like + Retweet if you enjoy this 🚫 Disclaimer *This information is not financial advice, and you are always responsible for the investment. **The above information can be changed to reference materials, and the provided information and actual information may be different, so please check the fact yourself.

@BIGBULLapp Multi agent panes sound small until you are juggling logs, approvals, and browser state. The UX win is keeping the loop visible.

Running Claude Code and Codex on Windows usually means firing up WSL and praying the shell does not eat your path variables. openwong2kim/wmux gives you split panes for multiple AI agents running side by side, no WSL required. It also ships an MCP bridge for browser automation, so the agents can actually click around instead of just hallucinating DOM selectors. Keep this bookmarked for the next time you need to run two models against the same task without leaving native Windows.