科技推特每日精选 - 2026-05-28

科技每日简报 (2026-05-28)

Today's top tech conversations are led by @Polymarket, whose post about 'JUST IN: Robinhood is rolling ...' garnered the highest engagement. Key themes trending across the top stories include https, product, revenue, claude, every. The community is actively discussing recent developments in AI, engineering practices, and startup strategies.

1. Polymarket (Group Score: 362.9 | Individual: 33.6)

Cluster: 17 tweets | Engagement: 2564 (Avg: 1710) | Type: Tech

JUST IN: Robinhood is rolling out AI agents that can trade stocks & make credit card purchases on users’ behalf via tools like Claude.

See 16 related tweets

@RoundtableSpace: Me on the phone with my credit card company explaining the charges made were from my ai agent:

htt...

@Cointelegraph: 🔥 JUST IN: Robinhood now lets AI agents trade stocks and make purchases for users. https://t.co/uNxf...
@zerohedge: Robinhood Lets Customers Use AI to Trade Stocks, Make Credit-Card Purchases -- WSJ

this will have s...

@arvidkahl: I know this is probably great for some, but it also is "what could possibly go wrong" at scale.

Can...

@Forbes: Stock trading app Robinhood will allow customers to use AI agents on the platform, the company said,...

2. harvey (Group Score: 259.3 | Individual: 37.4)

Cluster: 8 tweets | Engagement: 81 (Avg: 54) | Type: Tech

We're proud to partner with @trajectorylabs.

Legal AI should learn from the work lawyers already do: every redline should improve the next draft.

With Trajectory, we’re bringing continual learning to Harvey’s agents. https://t.co/tvQFN6Y1pk\n\nQT @rronak_: Today, @MichaelElabd, @QuantumArjun, and I are excited to announce Trajectory.

We are a research lab and product company building the platform for Continual Learning.

Our platform unlocks the signal already sitting in product usage, so companies can continuously post-train large-scale agentic models that outperform the frontier. @trajectorylabs

We’ve raised $15M from @Conviction, @BessemerVP, @radicalvcfund, @jeffdean, @drfeifei and more.

We’re partnering with some of the best AI-native companies: @ClayRunHQ @Harvey, @DecagonAI, @mercor_ai, @RogoAI to power their agentic systems, some of which we are already in production with.

We’ve brought together a world class research team from DeepMind, OpenAI, Apple, Meta Superintelligence, Amazon AGI, Scale AI, and an elite product team from Stripe and Figma.

AI will never again start on day one. Every correction, every retry, every edit will make products smarter. This is Continual Learning.

See 7 related tweets

@samdblond: RT @rronak_: Today, @MichaelElabd, @QuantumArjun, and I are excited to announce Trajectory.

We are ...

@baseten: Excited to support the @trajectorylabs team as their inference partner.

Powering live, continuous ...

@natolambert: The most likely way continual learning manifests in the coming few years is through products used di...
@Scobleizer: New trend: continual learning.

I first showed you this with Levangie Labs from @blevlabs.

You "gr...

@rohanpaul_ai: Cracking continual learning would make AI far more capable, because models could improve from real u...

3. wallstengine (Group Score: 170.1 | Individual: 36.3)

Cluster: 5 tweets | Engagement: 151 (Avg: 130) | Type: Tech

$SNOW Q1’27 EARNINGS HIGHLIGHTS

🔹 Revenue: $1.39B (Est.$ 1.32B) 🟢; +33% YoY 🔹 Adj. EPS: $0.39 (Est.$ 0.32) 🟢 🔹 Product Revenue: $1.33B; +34% YoY 🔹 RPO:$ 9.21B; +38% YoY 🔸 AWS Agreement: $6B multi-year agreement

Q2 Guide: 🔹 Product Revenue: $1.415B-$ 1.420B; +30% YoY 🔹 Non-GAAP Operating Margin: 12.5%

FY Guide: 🔹 Product Revenue: $5.840B; +31% YoY, up from$ 5.660B 🔹 Non-GAAP Product Gross Margin: 75.0% 🔹 Non-GAAP Operating Margin: 13.5%, up from prior 12.5% 🔹 Non-GAAP Adjusted FCF Margin: 23.0%

Other Metrics: 🔹 Net Revenue Retention: 126% 🔹 Customers With TTM Product Revenue >$1M: 779; +29% YoY 🔹 Forbes Global 2000 Customers: 813 🔹 Snowflake AI Accounts: 13,600+ 🔹 Cortex Code Accounts: 7,100+ 🔹 Net New Customers: 616; +38% YoY 🔹 New Forbes Global 2000 Customers: 13 🔹 Intent to Acquire: Natoma, an enterprise Model Context Protocol platform for AI agents

Financials: 🔹 Non-GAAP Product Gross Margin: 75.1% 🔹 Non-GAAP Operating Income: $165.8M 🔹 Non-GAAP Operating Margin: 11.9% 🔹 Operating Cash Flow:$ 243.2M 🔹 Adjusted FCF: $265.5M

Commentary: 🔸 “Snowflake delivered a milestone quarter, with product revenue of $1.33 billion, up 34% year-over-year, marking the strongest sequential dollar growth in our history.”

🔸 “AI continues to be a powerful tailwind for Snowflake, and Q1 marks a clear inflection point in that journey.”

🔸 “Given the strong momentum across both our core platform business and AI business, we are raising our full-year product revenue guidance.”

🔸 “Agents don’t just need access to data. They need the right context, permissions and policy guardrails to operate safely inside the enterprise.”

See 4 related tweets

@wallstengine: $MRVL Q1’27 EARNINGS HIGHLIGHTS

🔹 Revenue: $2.42B (Est.$ 2.40B) 🟢; +28% YoY 🔹 Adj. EPS: $0.80 (Est....

@wallstengine: $NTNX Q3’26 EARNINGS HIGHLIGHTS

🔹 Revenue: $703.1M (Est.$ 686.34M) 🟢; +10% YoY 🔹 Adj. EPS: $0.47 (E...

@wallstengine: $NCNO Q1’27 EARNINGS HIGHLIGHTS

🔹 Revenue: $159.4M (Est.$ 155.76M) 🟢; +11% YoY 🔹 Adj. EPS: $0.12 🔹...

@PatrickMoorhead: $MRVL had a great Q and raised the FY27 and FY28 outlook.

Revenue $2.42B, up 28% YoY, non-GAAP EPS ...

4. Reuters (Group Score: 166.8 | Individual: 28.4)

Cluster: 11 tweets | Engagement: 117 (Avg: 110) | Type: Tech

Nvidia plans to invest around $150 billion a year in Taiwan, CEO Jensen Huang said at the launch celebration for the chipmaker's planned Taiwan headquarters in Taipei, a week after rival AMD said it would invest more than$ 10 billion in Taiwan's AI sector https://t.co/EmEc6bSM7Y https://t.co/aJfZAM2zf2

See 10 related tweets

@Reuters: Nvidia to spend $150 billion a year in Taiwan, 'epicentre' of AI revolution, says CEO https://t.co/R...
@MTSlive: SITUATION DETECTED: Nvidia CEO Jensen Huang says the company plans to invest $150B a year in Taiwan,...
@StockSavvyShay: $NVDA unveiled its new Constellation AI campus in Taipei which is a 4,000-employee R&D hub expec...
@exec_sum: BREAKING: Nvidia plans to spend $150 billion per year in Taiwan

The firm will also build a Taiwan H...

@rohanpaul_ai: Jensen Huang presents new Taiwan campus.

Nvidia plans to invest around $150 bn a year in Taiwan,

...

5. ericschmidt (Group Score: 151.3 | Individual: 33.6)

Cluster: 5 tweets | Engagement: 52 (Avg: 52) | Type: Tech

well worth watching. Much progress in the best vision model\n\nQT @gs_ai_: We are back again :) After three weeks of quiet building.

Introducing Genesis World 1.0, our latest simulation platform, the second release in our full-stack suite. Open-sourced.

Robotics is still bottlenecked by the 1× speed of the physical world. Every model, checkpoint, and data recipe eventually needs to be tested on physical hardware, slowly, expensively, and with limited coverage.

One hour in reality can become 100 days in simulation. That is how robotics model iteration moves from a wall-clock bottleneck to a compute problem.

To make this work, simulation has to be both fast and trustworthy.

Over the past year, we rebuilt the entire stack: a GPU-accelerated cross-platform compiler, penetration-free multi-physics contact solvers, unified rigid and deformable physics, and a photo-realistic renderer purpose-built for physical AI applications.

We built Nyx, a high-performance path-traced rendering engine for robotics application.

Genesis World 1.0 achieves near realtime performance with our latest development for penetration-free IPC solver, supporting various types of deformables beyond rigid bodies. It supports contact-rich, dexterous manipulation simulation across different embodiments: unitree, sharpa, wuji, genesis hand and various types of grippers.

Under the hood is Quadrants, our effort in pushing forward cross-platform GPU-accelerated computation. Quadrants started as a fork of Taichi, and we rebuilt most of the critical parts for optimizing simulation workloads, giving 10x faster launch time and up to 4.6x runtime performance compared to the initial Genesis release.

Together, they bring us to an unprecedentedly low sim-to-real gap, enabling zero-shot real-to-sim model evaluation and much faster iteration of GENE.

All available today. Genesis World 1.0: https://t.co/aknCM3eqws Quadrants: https://t.co/uXqPNI4cb6 Nyx: https://t.co/R8j0djqGnV

See 4 related tweets

@chris_j_paxton: Looks like a really nice simulator\n\nQT @gs_ai_: We are back again :) After three weeks of quiet bu...
@Scobleizer: Robot builders have amazing new tools.\n\nQT @gs_ai_: We are back again :) After three weeks of quie...
@vkhosla: Cool gets cooler\n\nQT @gs_ai_: We are back again :) After three weeks of quiet building.

Introduci...

@gs_ai_: RT @lukas_m_ziegler: That's sick! 🤯

Genesis AI simulates robots playing yo-yo! 🪀

@gs_ai_ just open...

6. Parul_Gautam7 (Group Score: 150.4 | Individual: 36.9)

Cluster: 5 tweets | Engagement: 50 (Avg: 95) | Type: Tech

Genspark AI Slides 5.0 feels like a real shift from AI-generated presentations to AI-powered expert thinking.

The whole Slide Skills concept makes expertise reusable, scalable, and instantly accessible.

That’s the kind of upgrade that can genuinely change how teams build decks.\n\nQT @genspark_ai: Introducing Genspark AI Slides 5.0. Our biggest upgrade to AI Slides yet.

The biggest leap? Slide Skills. Reusable assets that bundle expert thinking and polished design, ready in one click.

Here's everything new in 5.0:

100+ Expert Skills built in. We've packed the thinking of the world's top experts, across consulting, sales, teaching, and more. Apply with one click, and their expertise becomes yours.
Build your own Skill. Upload a deck or PDF, describe the analysis steps, and lock in your team's style.
5× faster, higher quality. Powered by next-gen architecture and the most powerful models.
Built-in data analysis. Crunches your Excel and Sheets data into presentation-ready insights.

Every deck thinks like an expert.

Try it now → https://t.co/VSPHzheU7X

🎉 50% off AI Slides 5.0 credits, 14 days only

See 4 related tweets

@rohanpaul_ai: Genspark AI Slides It's a beautiful piece of tech for so many office work situations.

They just up...

@dani_avila7: slide skills look interesting 👀

reusable expert templates I can apply to any deck

going to start b...

@sang_wen: RT @genspark_ai: Introducing Genspark AI Slides 5.0. Our biggest upgrade to AI Slides yet.

The bigg...

@Parul_Gautam7: RT @Parul_Gautam7: Genspark AI Slides 5.0 feels like a real shift from AI-generated presentations to...

7. dee_bosa (Group Score: 140.0 | Individual: 32.9)

Cluster: 6 tweets | Engagement: 103 (Avg: 167) | Type: Tech

Higher spend will create a new software category. Just like AWS bills didn't kill cloud, they created FinOps.

enterprise needs a way to track which teams/models/workflows drive spend and route simple tasks to cheaper models and measure quality per dollar.

Already a cohort of companies here: Ramp, Databricks, Factory AI. Who else?\n\nQT @arakharazian: The average business is spending 13x more on AI tokens than in January 2025, according to @tryramp data.

That’s driven growth for OpenAI and Anthropic. While tokens are still a small share of actual business spend (<2% even for the highest spenders), this trend is unsustainable for most businesses.

Two players will win out as businesses pivot to cost discipline on AI:

— Google and Microsoft, who can more competitively offer cheaper models (they’ll be profitable anyway) — Platforms (Cursor, OpenRouter) that offer access to smart routing and open source models, cheaper than what you get from the main model companies.

My comments in @dealbook today.

See 5 related tweets

@bradmenezes: Tokenmaxxing was cool for a hot minute.

Now the pendulum is swinging back hard.

When Uber’s CT...

@Mayhem4Markets: Remember when OpenAI was actually supposed to be a non-profit whose original purpose was to fight ag...
@TheStalwart: Feels like optimal query-routing technology is going to get very hot.\n\nQT @arakharazian: The avera...
@arakharazian: RT @dee_bosa: Higher spend will create a new software category. Just like AWS bills didn't kill clou...
@dee_bosa: RT @arakharazian: The average business is spending 13x more on AI tokens than in January 2025, accor...

8. VaibhavSisinty (Group Score: 137.3 | Individual: 31.5)

Cluster: 5 tweets | Engagement: 40 (Avg: 98) | Type: Tech

Anthropic just shipped the same security tool they use internally for free, inside Claude Code. 😨

It catches vulnerabilities as they're being written. Three levels of review:

→ On every file edit: flags risky code patterns and misused libraries

→ After Claude finishes a chunk of work: reviews the full diff for issues that only show up across multiple changes

→ On commits: reads the surrounding code to confirm if a vulnerability is real (cuts false alarms)

You can also drop a claude-security-guidance.md file in your repo to enforce your company's own security rules alongside the built-in ones.

For context: Anthropic has been running this on their own code internally. Security-related comments on pull requests dropped by 30 to 40%.

Before this a human reviewers used to flag ~100 security issues per code review. Now they flag 60 to 70. The plugin catches the rest before a human ever has to look.

Available now. Install from /plugins inside Claude Code.\n\nQT @ClaudeDevs: We’ve shipped a security-guidance plugin for Claude Code that helps identify and fix vulnerabilities as you’re writing code.

Available for all Claude Code users. Install from the plugin marketplace (/plugins). https://t.co/LprgC4m6Kf

See 4 related tweets

@WesRoth: Anthropic released a new security-guidance plugin for Claude Code that helps developers identify and...
@RoundtableSpace: ANTHROPIC JUST SHIPPED A SECURITY PLUGIN FOR CLAUDE CODE THAT IDENTIFIES AND FIXES VULNERABILITIES I...
@bcherny: RT @ClaudeDevs: We’ve shipped a security-guidance plugin for Claude Code that helps identify and fix...
@RoundtableSpace: CLAUDE CODE RELEASED A NEW SECURITY GUIDANCE PLUGIN THAT HELPS DEVELOPERS IDENTIFY AND FIX VULNERABI...

9. cgtwts (Group Score: 135.5 | Individual: 48.8)

Cluster: 5 tweets | Engagement: 6052 (Avg: 551) | Type: Tech

Anthropic CEO right now: https://t.co/B2PhKUczZj\n\nQT @Polymarket: JUST IN: Sam Altman says AI probably won’t trigger the “jobs apocalypse” he once predicted.

See 4 related tweets

@HedgieMarkets: 🦔Sam Altman told the Commonwealth Bank of Australia conference in Sydney yesterday that the AI "jobs...
@ShanuMathew93: Anthropic comms every time a job loss quote hits the air waves https://t.co/yY83EgMWAK\n\nQT @unusua...
@Cointelegraph: ⚡️ NEW: OpenAI CEO Sam Altman says AI is unlikely to cause a "jobs apocalypse," adding white-collar ...
@exec_sum: Sam Altman said AI is unlikely to lead to a 'jobs apocalypse'

"I'm delighted to ⁠be wrong about thi...

10. myrhex (Group Score: 131.5 | Individual: 30.1)

Cluster: 6 tweets | Engagement: 22 (Avg: 70) | Type: Tech

SpaceXAI has added another strong integration.

You can now use your SuperGrok or X Premium+ subscription directly inside @kilocode.

This gives you access to grok-build-0.1 for high-speed agentic coding intelligence, available in the Kilo IDE extensions or CLI.

The team continues to make Grok more useful across different developer tools and environments. Another practical step forward.

Link: https://t.co/MaHp6A5AT9\n\nQT @xai: Use your SuperGrok or X Premium+ subscription in @kilocode.

Try grok-build-0.1 for high speed and agentic coding intelligence, available in the Kilo IDE extensions or CLI.

https://t.co/JGO8UQKKEB

See 5 related tweets

@XFreeze: Starting today, you can now also use your 𝕏 Premium+ or SuperGrok subscription directly inside Kilo ...
@myrhex: Grok Build has just been updated to version 0.2.4.

The pace of development from SpaceXAI is becomin...

@xai: Use your SuperGrok or X Premium+ subscription in @kilocode.

Try grok-build-0.1 for high speed and ...

@teslaownersSV: NEWS: Kilo Code now supports direct access to Grok models through 𝕏 Premium+ and SuperGrok subscript...
@MarioNawfal: RT @MarioNawfal: 🇺🇸 xAI launched Grok Build, a coding agent that runs directly from your terminal, n...

11. WesRoth (Group Score: 119.7 | Individual: 31.1)

Cluster: 5 tweets | Engagement: 51 (Avg: 26) | Type: Tech

DeepSWE was released as a new benchmark for evaluating agentic coding models on more realistic software engineering tasks.

The benchmark is designed to show where top coding agents actually diverge in day-to-day developer workflows, even when public leaderboards make them look close in performance.\n\nQT @serenaa_ge: Today we’re releasing DeepSWE, a new standard for agentic coding benchmarks.

On public leaderboards, top models often look relatively close in capability. DeepSWE shows where they actually diverge, reflecting the realistic experience of developers in their day-to-day work. https://t.co/HCDcjNuTFK

See 4 related tweets

@QuixiAI: RT @serenaa_ge: Today we’re releasing DeepSWE, a new standard for agentic coding benchmarks.

On pub...

@Winterrose: incredible agentic coding benchmark shipped by the @serenaa_ge and the @datacurve team\n\nQT @serena...
@OfirPress: mini-swe-agent is our ~150 line of code agent, and it performs as well as Claude Code, Codex and Gem...
@andersonbcdefg: official confirmation that the claude code harness has become slop btw https://t.co/FwU3dGOyrz\n\nQT...

12. ajambrosino (Group Score: 118.9 | Individual: 36.0)

Cluster: 4 tweets | Engagement: 203 (Avg: 160) | Type: Tech

landed another 🐐👀\n\nQT @pvncher: Yesterday was my first day at @OpenAI working under the amazing @romainhuet.

I've spent the better part of two years pushing the frontier of having models write code for you, shipping updates week after week for @RepoPrompt.

Transitioning from being a founder is never easy, especially when you have a community of amazing developers who invested so much in your tool.

Thankfully Romain worked hard to ensure that all those users would be taken care in the process, and if you're one such user, you should have an email waiting in your inbox with the details!

I'm very excited to be joining this talented team and work alongside everyone at OpenAI contributing to codex.

See 3 related tweets

@romainhuet: Very excited to welcome @pvncher to OpenAI!

Eric built @RepoPrompt with a lot of taste and care. We...

@ivanfioravanti: RT @pvncher: Yesterday was my first day at @OpenAI working under the amazing @romainhuet.

I've spen...

@Dimillian: RT @romainhuet: Very excited to welcome @pvncher to OpenAI!

Eric built @RepoPrompt with a lot of ta...

13. MarioNawfal (Group Score: 115.3 | Individual: 33.0)

Cluster: 4 tweets | Engagement: 217 (Avg: 571) | Type: Tech

Built a full website in under 60 minutes using Grok Build. Insane.

One prompt: “Create a website for my whale watching tours business in Hermanus… generate images…”

Grok Build just runs. You answer a few questions in the meantime. Voilà.

What Grok Build actually does:

Creates the full project; folders, structure, ready to go
Generates custom images; no stock, no uploads
Builds the site; layout, sections, pricing
Adds features; gallery, booking form, interactions
Writes styling; responsive, clean, animated
Handles logic; menus, forms, smooth UX
Runs it locally; live and testable instantly

No dragging blocks. No templates. Just built.

This is execution, not assistance.

@xAI\n\nQT @elonmusk: How to use Grok Build

See 3 related tweets

@WesRoth: Grok is reportedly preparing a dedicated Build tab on Grok Web, creating a direct entry point for Gr...
@MarioNawfal: RT @MarioNawfal: Built a full website in under 60 minutes using Grok Build. Insane.

One prompt: “Cr...

@MarioNawfal: Grok Build just made web dev feel like cheating.

A user built a full website covering every war in ...

14. business (Group Score: 115.1 | Individual: 49.4)

Cluster: 4 tweets | Engagement: 680 (Avg: 80) | Type: Tech

The founder of failed Indian technology firm Byju’s was sentenced to six months in jail by a Singapore court for contempt, sources say https://t.co/rNSJbSXSha

See 3 related tweets

@business: The founder of failed Indian technology firm Byju’s was sentenced to six months in jail by a Singapo...
@Polymarket: JUST IN: Founder of failed Indian startup Byju’s sentenced to six months in jail for contempt of cou...
@exec_sum: Byju Raveendran, the once-billionaire founder of Indian edtech BYJU'S, was sentenced to six months i...

15. wallstengine (Group Score: 113.5 | Individual: 32.8)

Cluster: 4 tweets | Engagement: 36 (Avg: 130) | Type: Tech

Farsight AI is launching Freeform, an AI agent for financial services teams that can create full client-ready deliverables from one prompt.

The product is aimed at long-form finance work like pitch decks, CIMs, IC memos, financial models, and Word docs.\n\nQT @TangriKunal: Most AI-generated materials optimize for a generic sense of "good looking". Today we're launching Freeform, built to understand how "good" differs from one client to the next.

We've spent 3 years at @farsight_ai working on deliverables creation for financial services. The first years were spent making output look polished. But polished for one client means something completely different from another, and that's a taste problem most AI doesn't solve for.

Freeform does.

One prompt. Full 80-page pitch decks, memos, and financial models, ready to send to clients, in under 30 minutes across PowerPoint, Excel, and Word.

See 3 related tweets

@Mayhem4Markets: You spend Fridays fixing slide fonts and reconciling Excel links.

Freeform handles the production. ...

@shiri_shh: financial analysts watching this cook their entire career path live 😭 https://t.co/S1gx7IdkjL\n\nQT ...
@cgtwts: Finance analysts right now: https://t.co/jiWh9Z4jS7\n\nQT @TangriKunal: Most AI-generated materials ...

16. business (Group Score: 113.0 | Individual: 25.9)

Cluster: 6 tweets | Engagement: 190 (Avg: 80) | Type: Tech

Cognition AI has raised more than $1 billion in a new funding round at a$ 26 billion valuation. https://t.co/BTIIauX347

See 5 related tweets

@Techmeme: AI coding startup Cognition AI raised more than $1B at a$ 26B valuation, and says its revenue run ra...
@MTSlive: SITUATION DETECTED: Cognition AI has raised more than $1B at a$ 26B valuation, per Bloomberg.

Lux C...

@TechCrunch: AI coding startup Cognition raises $1B at$ 25B pre-money valuation https://t.co/5Lrkm28aFC...
@techdaily24: AI coding startup Cognition raises $1B at$ 25B pre-money valuation https://t.co/g8hTIYyrw2 #technolo...
@FirstSquawk: AI CODING COMPANY COGNITION SECURES $1 BILLION FUNDING WITH A VALUATION OF$ 26 BILLION....

17. PatrickMoorhead (Group Score: 110.4 | Individual: 30.4)

Cluster: 5 tweets | Engagement: 48 (Avg: 32) | Type: Tech

Most people receive gifts on their birthday, but not @danielnewmanUV, who bought a company (ETR). I can't wait to see what he does with it.\n\nQT @danielnewmanUV: Wall Street already runs on ETR's signal. Today, Futurum signed a definitive agreement to acquire the dataset institutional investors trust to read enterprise tech spend.

Here is why it matters.

→ 9,000+ enterprise IT decision-makers feeding real spending intent, not vendor-sourced guesses

→ $2 Trillion in tracked technology spend that hedge funds and institutions use to call the print before the print

→ The earliest read on who is winning and losing budget across every major name in tech

→ Now sitting on the Futurum Intelligence Platform, paired with our analysts, our coverage, and our reach

The legacy research firms tell you what happened last quarter. This data tells you what is happening right now.

Research built on evidence. This is what AI-native looks like when it has the data to back it up.

Futurum and ETR. Subject to customary closing conditions.

See 4 related tweets

@TheFuturumGroup: The future of market intelligence depends on data that is deeper, faster, and more connected to real...
@StockSavvyShay: RT @danielnewmanUV: Wall Street already runs on ETR's signal. Today, Futurum signed a definitive agr...
@TheFuturumGroup: Better decisions require better signal.

@TheFuturumGroup has entered into a definitive agreement to...

@Signal_65: RT @TheFuturumGroup: 🤝 @TheFuturumGroup has entered into a definitive agreement to acquire @ETRnews....

18. Docker (Group Score: 108.2 | Individual: 29.3)

Cluster: 4 tweets | Engagement: 8 (Avg: 54) | Type: Tech

AI is putting infrastructure back at the center.

Proud to see Docker included in @Redpoint's 2026 InfraRed 100 - recognizing the companies that are helping define the next phase of cloud and AI infrastructure.\n\nQT @Redpoint: The Redpoint InfraRed 100 is now live.

These are the companies building the infrastructure that powers everything happening in AI right now, from world models and agent runtimes to the sandboxes, databases, and security tools agents depend on.

Congratulations to this year's honorees!

Read the full 2026 InfraRed Report: our state of the union on AI and cloud infrastructure 👉 https://t.co/Y1y94ZwI5B

See 3 related tweets

@ClickHouseDB: Honored to be named in @Redpoint 's 2026 InfraRed 100 alongside so many innovative infrastructure an...
@Replit: RT @Redpoint: The Redpoint InfraRed 100 is now live.

These are the companies building the infrastru...

@glean: Thrilled to be named to @Redpoint’s 2026 InfraRed 100!

AI transformation needs more than models; it...

19. dhaber (Group Score: 106.8 | Individual: 40.2)

Cluster: 3 tweets | Engagement: 158 (Avg: 69) | Type: Tech

Marc Rowan has quietly built Apollo from a distressed-focused private equity firm into one of the most important financial institutions in the world, now managing over $1T in assets and becoming the largest provider of retirement income globally.

In this fantastic conversation we covered lessons at Drexel from Michael Milken, Apollo’s origin story from a cold call from France, why AI could outgrow venture equity, and Apollo’s “play to win” culture.

We also connected over our shared belief that “Opportunities live between fields of expertise.”

This was really fun, hope you enjoy! cc @apolloglobal\n\nQT @a16z: Financing the Global Industrial Renaissance with Apollo CEO Marc Rowan

Marc Rowan is CEO of Apollo Global Management, one of the most important financial institutions and the largest provider of retirement income in the world. In this conversation, he joins a16z's David Haber to discuss the story of Apollo, the state of public and private markets, how the AI revolution will be financed, and more.

00:00 Intro 00:52 Drexel, Milken & the origins of "clean sheet thinking" 04:55 The Apollo origin story: From unemployed to $6 billion 08:46 How Apollo became a trillion-dollar firm 13:00 Permanent capital, origination & why assets are the scarce resource 16:08 Democratizing private markets: Daily pricing & new capital channels 22:04 Where venture meets credit: Financing the industrial renaissance 30:01 AI, enterprise software & which jobs will be replaced or enhanced 38:52 Moral leadership: UPenn, merit & doing right over easy 46:02 Apollo's culture: Playing to win & building to outlast the founder

See 2 related tweets

@a16z: Financing the Global Industrial Renaissance with Apollo CEO Marc Rowan

Marc Rowan is CEO of Apollo ...

@dhaber: "Every dollar since the invention of fire" is going into this AI capex build out — and equity can’t...

20. OpenBMB (Group Score: 103.5 | Individual: 34.3)

Cluster: 4 tweets | Engagement: 47 (Avg: 74) | Type: Tech

A huge thank you to the team at @ArtificialAnlys for recognizing MiniCPM5-1B in your Intelligence Index! It’s the highest-scoring open weights model under 2B parameters on the AA Intelligence Index — and does so with 31x fewer tokens than the next best.\n\nQT @ArtificialAnlys: OpenBMB has released MiniCPM5-1B (Non-reasoning), the leading 1B open weights model, scoring 17.9 on the Artificial Analysis Intelligence Index

@OpenBMB is a China-based lab jointly founded in 2022 by Tsinghua University’s NLP Lab and ModelBest Inc. This release extends the open weights Pareto frontier for Intelligence vs. Parameters at the sub-2B scale. It sits almost 2 points ahead of the best-performing 2B open weights model, @Alibaba's Qwen3.5 2B (Reasoning, 16.3), and 7 points ahead of Qwen3.5 0.8B (Reasoning, 10.5).

Unlike the recently released MiniCPM-V 4.6 1.3B Instruct, MiniCPM5-1B (Non-reasoning) does not support native multimodal input, and is text input and output only.

Key results:

➤ MiniCPM5-1B scores 17.9 on the Artificial Analysis Intelligence Index, the highest of any open weights model at 1B parameters or below by 7.4 points. The next-most-intelligent open weights model at this scale is Qwen3.5 0.8B (Reasoning, 10.5). No other open weights model under 2B parameters has exceeded 15 on the Intelligence Index; its predecessor MiniCPM-V 4.6 1.3B sits at 12.7.

➤ MiniCPM5-1B extends the open weights Pareto frontier on both Intelligence vs. Total Parameters and Intelligence vs. Active Parameters at the sub-2B scale. It surpasses its predecessor MiniCPM-V 4.6 1.3B (12.7) by 5.3 points at ~23% fewer parameters, and beats Qwen3.5 2B (Reasoning, 16.3) by 1.6 points at less than half the parameter count.

➤ MiniCPM5-1B is more token-efficient than the larger reasoning peers it surpasses, but uses more output tokens than its (also non-reasoning) predecessor MiniCPM-V 4.6 1.3B. It used 12.6M output tokens to run the Intelligence Index, ~31x fewer than Qwen3.5 2B (Reasoning, 389M) and ~8x fewer than Qwen3.5 2B (Non-reasoning, 100M), but ~2.3x more than MiniCPM-V 4.6 1.3B's 5.4M.

➤ AA-Omniscience score of -1 is the highest in its size class, earned by abstaining rather than hallucinating. MiniCPM5-1B declines to answer the vast majority of AA-Omniscience questions, avoiding the hallucination penalty that pulls sub-2B peers down to the -70 to -89 range (Qwen3.5 0.8B Non-reasoning at -89, MiniCPM-V 4.6 1.3B at -85, Exaone 4.0 1.2B Non-reasoning at -83). Choosing to abstain rather than guess is the more honest posture, and AA-Omniscience credits it positively.

Additional model details:

➤ Size: 1B total parameters (dense)

➤ Context window: 128K

➤ Modality: Text input and output only

➤ Precision: BF16

➤ License: Apache 2.0

➤ Providers: No confirmed providers upon release

See 3 related tweets

@TeksEdge: A new model to try out on @LocallyAIApp for the iPhone. Currently running Bonsai 8B but MiniCMP5-1B...
@OpenBMB: RT @ArtificialAnlys: OpenBMB has released MiniCPM5-1B (Non-reasoning), the leading 1B open weights m...
@teortaxesTex: RT @OpenBMB: A huge thank you to the team at @ArtificialAnlys for recognizing MiniCPM5-1B in your In...